diff --git a/tests/snapshots/gemma-3-4b-it.schema b/tests/snapshots/gemma-3-4b-it.schema index cf5032a24d..cb8718a889 100644 --- a/tests/snapshots/gemma-3-4b-it.schema +++ b/tests/snapshots/gemma-3-4b-it.schema @@ -21,433 +21,433 @@ token_embd.weight q6_K [Q2_K] q2_K token_embd.weight q6_K -blk.0.ffn_down.weight q3_K blk.0.attn_output.weight q3_K blk.0.attn_v.weight q3_K -blk.1.ffn_down.weight q3_K +blk.0.ffn_down.weight q3_K blk.1.attn_output.weight q3_K blk.1.attn_v.weight q3_K -blk.10.ffn_down.weight q3_K -blk.10.attn_output.weight q3_K -blk.10.attn_v.weight q3_K -blk.11.ffn_down.weight q3_K -blk.11.attn_output.weight q3_K -blk.11.attn_v.weight q3_K -blk.12.ffn_down.weight q3_K -blk.12.attn_output.weight q3_K -blk.12.attn_v.weight q3_K -blk.13.ffn_down.weight q3_K -blk.13.attn_output.weight q3_K -blk.13.attn_v.weight q3_K -blk.14.attn_output.weight q3_K -blk.14.attn_v.weight q3_K -blk.2.ffn_down.weight q3_K +blk.1.ffn_down.weight q3_K blk.2.attn_output.weight q3_K blk.2.attn_v.weight q3_K -blk.3.ffn_down.weight q3_K +blk.2.ffn_down.weight q3_K blk.3.attn_output.weight q3_K blk.3.attn_v.weight q3_K -blk.4.ffn_down.weight q3_K +blk.3.ffn_down.weight q3_K blk.4.attn_output.weight q3_K blk.4.attn_v.weight q3_K -blk.5.ffn_down.weight q3_K +blk.4.ffn_down.weight q3_K blk.5.attn_output.weight q3_K blk.5.attn_v.weight q3_K -blk.6.ffn_down.weight q3_K +blk.5.ffn_down.weight q3_K blk.6.attn_output.weight q3_K blk.6.attn_v.weight q3_K -blk.7.ffn_down.weight q3_K +blk.6.ffn_down.weight q3_K blk.7.attn_output.weight q3_K blk.7.attn_v.weight q3_K -blk.8.ffn_down.weight q3_K +blk.7.ffn_down.weight q3_K blk.8.attn_output.weight q3_K blk.8.attn_v.weight q3_K -blk.9.ffn_down.weight q3_K +blk.8.ffn_down.weight q3_K blk.9.attn_output.weight q3_K blk.9.attn_v.weight q3_K +blk.9.ffn_down.weight q3_K +blk.10.attn_output.weight q3_K +blk.10.attn_v.weight q3_K +blk.10.ffn_down.weight q3_K +blk.11.attn_output.weight q3_K +blk.11.attn_v.weight q3_K +blk.11.ffn_down.weight q3_K +blk.12.attn_output.weight q3_K +blk.12.attn_v.weight q3_K +blk.12.ffn_down.weight q3_K +blk.13.attn_output.weight q3_K +blk.13.attn_v.weight q3_K +blk.13.ffn_down.weight q3_K +blk.14.attn_output.weight q3_K +blk.14.attn_v.weight q3_K blk.14.ffn_down.weight q3_K -blk.15.ffn_down.weight q3_K blk.15.attn_output.weight q3_K blk.15.attn_v.weight q3_K -blk.16.ffn_down.weight q3_K +blk.15.ffn_down.weight q3_K blk.16.attn_output.weight q3_K blk.16.attn_v.weight q3_K -blk.17.ffn_down.weight q3_K +blk.16.ffn_down.weight q3_K blk.17.attn_output.weight q3_K blk.17.attn_v.weight q3_K -blk.18.ffn_down.weight q3_K +blk.17.ffn_down.weight q3_K blk.18.attn_output.weight q3_K blk.18.attn_v.weight q3_K -blk.19.ffn_down.weight q3_K +blk.18.ffn_down.weight q3_K blk.19.attn_output.weight q3_K blk.19.attn_v.weight q3_K -blk.20.ffn_down.weight q3_K +blk.19.ffn_down.weight q3_K blk.20.attn_output.weight q3_K blk.20.attn_v.weight q3_K -blk.21.ffn_down.weight q3_K +blk.20.ffn_down.weight q3_K blk.21.attn_output.weight q3_K blk.21.attn_v.weight q3_K -blk.22.ffn_down.weight q3_K +blk.21.ffn_down.weight q3_K blk.22.attn_output.weight q3_K blk.22.attn_v.weight q3_K -blk.23.ffn_down.weight q3_K +blk.22.ffn_down.weight q3_K blk.23.attn_output.weight q3_K blk.23.attn_v.weight q3_K -blk.24.ffn_down.weight q3_K +blk.23.ffn_down.weight q3_K blk.24.attn_output.weight q3_K blk.24.attn_v.weight q3_K -blk.25.ffn_down.weight q3_K +blk.24.ffn_down.weight q3_K blk.25.attn_output.weight q3_K blk.25.attn_v.weight q3_K -blk.26.ffn_down.weight q3_K +blk.25.ffn_down.weight q3_K blk.26.attn_output.weight q3_K blk.26.attn_v.weight q3_K -blk.27.ffn_down.weight q3_K +blk.26.ffn_down.weight q3_K blk.27.attn_output.weight q3_K blk.27.attn_v.weight q3_K -blk.28.ffn_down.weight q3_K +blk.27.ffn_down.weight q3_K blk.28.attn_output.weight q3_K blk.28.attn_v.weight q3_K -blk.29.ffn_down.weight q3_K +blk.28.ffn_down.weight q3_K blk.29.attn_output.weight q3_K blk.29.attn_v.weight q3_K -blk.30.ffn_down.weight q3_K +blk.29.ffn_down.weight q3_K blk.30.attn_output.weight q3_K blk.30.attn_v.weight q3_K -blk.31.ffn_down.weight q3_K +blk.30.ffn_down.weight q3_K blk.31.attn_output.weight q3_K blk.31.attn_v.weight q3_K -blk.32.ffn_down.weight q3_K +blk.31.ffn_down.weight q3_K blk.32.attn_output.weight q3_K blk.32.attn_v.weight q3_K -blk.33.ffn_down.weight q3_K +blk.32.ffn_down.weight q3_K blk.33.attn_output.weight q3_K blk.33.attn_v.weight q3_K +blk.33.ffn_down.weight q3_K [Q3_K_S] q3_K token_embd.weight q6_K [Q3_K_M] q3_K token_embd.weight q6_K -blk.0.ffn_down.weight q5_K blk.0.attn_output.weight q4_K blk.0.attn_v.weight q5_K -blk.1.ffn_down.weight q5_K +blk.0.ffn_down.weight q5_K blk.1.attn_output.weight q4_K blk.1.attn_v.weight q5_K -blk.10.ffn_down.weight q4_K -blk.10.attn_output.weight q4_K -blk.10.attn_v.weight q4_K -blk.11.ffn_down.weight q4_K -blk.11.attn_output.weight q4_K -blk.11.attn_v.weight q4_K -blk.12.ffn_down.weight q4_K -blk.12.attn_output.weight q4_K -blk.12.attn_v.weight q4_K -blk.13.ffn_down.weight q4_K -blk.13.attn_output.weight q4_K -blk.13.attn_v.weight q4_K -blk.14.attn_output.weight q4_K -blk.14.attn_v.weight q4_K -blk.2.ffn_down.weight q4_K +blk.1.ffn_down.weight q5_K blk.2.attn_output.weight q4_K blk.2.attn_v.weight q4_K -blk.3.ffn_down.weight q4_K +blk.2.ffn_down.weight q4_K blk.3.attn_output.weight q4_K blk.3.attn_v.weight q4_K -blk.4.ffn_down.weight q4_K +blk.3.ffn_down.weight q4_K blk.4.attn_output.weight q4_K blk.4.attn_v.weight q4_K -blk.5.ffn_down.weight q4_K +blk.4.ffn_down.weight q4_K blk.5.attn_output.weight q4_K blk.5.attn_v.weight q4_K -blk.6.ffn_down.weight q4_K +blk.5.ffn_down.weight q4_K blk.6.attn_output.weight q4_K blk.6.attn_v.weight q4_K -blk.7.ffn_down.weight q4_K +blk.6.ffn_down.weight q4_K blk.7.attn_output.weight q4_K blk.7.attn_v.weight q4_K -blk.8.ffn_down.weight q4_K +blk.7.ffn_down.weight q4_K blk.8.attn_output.weight q4_K blk.8.attn_v.weight q4_K -blk.9.ffn_down.weight q4_K +blk.8.ffn_down.weight q4_K blk.9.attn_output.weight q4_K blk.9.attn_v.weight q4_K +blk.9.ffn_down.weight q4_K +blk.10.attn_output.weight q4_K +blk.10.attn_v.weight q4_K +blk.10.ffn_down.weight q4_K +blk.11.attn_output.weight q4_K +blk.11.attn_v.weight q4_K +blk.11.ffn_down.weight q4_K +blk.12.attn_output.weight q4_K +blk.12.attn_v.weight q4_K +blk.12.ffn_down.weight q4_K +blk.13.attn_output.weight q4_K +blk.13.attn_v.weight q4_K +blk.13.ffn_down.weight q4_K +blk.14.attn_output.weight q4_K +blk.14.attn_v.weight q4_K blk.14.ffn_down.weight q4_K -blk.15.ffn_down.weight q4_K blk.15.attn_output.weight q4_K blk.15.attn_v.weight q4_K -blk.16.ffn_down.weight q4_K +blk.15.ffn_down.weight q4_K blk.16.attn_output.weight q4_K blk.16.attn_v.weight q4_K -blk.17.ffn_down.weight q4_K +blk.16.ffn_down.weight q4_K blk.17.attn_output.weight q4_K blk.17.attn_v.weight q4_K -blk.18.ffn_down.weight q4_K +blk.17.ffn_down.weight q4_K blk.18.attn_output.weight q4_K blk.18.attn_v.weight q4_K -blk.19.ffn_down.weight q4_K +blk.18.ffn_down.weight q4_K blk.19.attn_output.weight q4_K blk.19.attn_v.weight q4_K -blk.20.ffn_down.weight q4_K +blk.19.ffn_down.weight q4_K blk.20.attn_output.weight q4_K blk.20.attn_v.weight q4_K -blk.21.ffn_down.weight q4_K +blk.20.ffn_down.weight q4_K blk.21.attn_output.weight q4_K blk.21.attn_v.weight q4_K -blk.22.ffn_down.weight q4_K +blk.21.ffn_down.weight q4_K blk.22.attn_output.weight q4_K blk.22.attn_v.weight q4_K -blk.23.ffn_down.weight q4_K +blk.22.ffn_down.weight q4_K blk.23.attn_output.weight q4_K blk.23.attn_v.weight q4_K -blk.24.ffn_down.weight q4_K +blk.23.ffn_down.weight q4_K blk.24.attn_output.weight q4_K blk.24.attn_v.weight q4_K -blk.25.ffn_down.weight q4_K +blk.24.ffn_down.weight q4_K blk.25.attn_output.weight q4_K blk.25.attn_v.weight q4_K -blk.26.ffn_down.weight q4_K +blk.25.ffn_down.weight q4_K blk.26.attn_output.weight q4_K blk.26.attn_v.weight q4_K -blk.27.ffn_down.weight q4_K +blk.26.ffn_down.weight q4_K blk.27.attn_output.weight q4_K blk.27.attn_v.weight q4_K -blk.28.ffn_down.weight q4_K +blk.27.ffn_down.weight q4_K blk.28.attn_output.weight q4_K blk.28.attn_v.weight q4_K -blk.29.ffn_down.weight q4_K +blk.28.ffn_down.weight q4_K blk.29.attn_output.weight q4_K blk.29.attn_v.weight q4_K -blk.30.ffn_down.weight q4_K +blk.29.ffn_down.weight q4_K blk.30.attn_output.weight q4_K blk.30.attn_v.weight q4_K -blk.31.ffn_down.weight q4_K +blk.30.ffn_down.weight q4_K blk.31.attn_output.weight q4_K blk.31.attn_v.weight q4_K -blk.32.ffn_down.weight q4_K +blk.31.ffn_down.weight q4_K blk.32.attn_output.weight q4_K blk.32.attn_v.weight q4_K -blk.33.ffn_down.weight q4_K +blk.32.ffn_down.weight q4_K blk.33.attn_output.weight q4_K blk.33.attn_v.weight q4_K +blk.33.ffn_down.weight q4_K [Q3_K_L] q3_K token_embd.weight q6_K -blk.0.ffn_down.weight q5_K blk.0.attn_output.weight q5_K blk.0.attn_v.weight q5_K -blk.1.ffn_down.weight q5_K +blk.0.ffn_down.weight q5_K blk.1.attn_output.weight q5_K blk.1.attn_v.weight q5_K -blk.10.ffn_down.weight q5_K -blk.10.attn_output.weight q5_K -blk.10.attn_v.weight q5_K -blk.11.ffn_down.weight q5_K -blk.11.attn_output.weight q5_K -blk.11.attn_v.weight q5_K -blk.12.ffn_down.weight q5_K -blk.12.attn_output.weight q5_K -blk.12.attn_v.weight q5_K -blk.13.ffn_down.weight q5_K -blk.13.attn_output.weight q5_K -blk.13.attn_v.weight q5_K -blk.14.attn_output.weight q5_K -blk.14.attn_v.weight q5_K -blk.2.ffn_down.weight q5_K +blk.1.ffn_down.weight q5_K blk.2.attn_output.weight q5_K blk.2.attn_v.weight q5_K -blk.3.ffn_down.weight q5_K +blk.2.ffn_down.weight q5_K blk.3.attn_output.weight q5_K blk.3.attn_v.weight q5_K -blk.4.ffn_down.weight q5_K +blk.3.ffn_down.weight q5_K blk.4.attn_output.weight q5_K blk.4.attn_v.weight q5_K -blk.5.ffn_down.weight q5_K +blk.4.ffn_down.weight q5_K blk.5.attn_output.weight q5_K blk.5.attn_v.weight q5_K -blk.6.ffn_down.weight q5_K +blk.5.ffn_down.weight q5_K blk.6.attn_output.weight q5_K blk.6.attn_v.weight q5_K -blk.7.ffn_down.weight q5_K +blk.6.ffn_down.weight q5_K blk.7.attn_output.weight q5_K blk.7.attn_v.weight q5_K -blk.8.ffn_down.weight q5_K +blk.7.ffn_down.weight q5_K blk.8.attn_output.weight q5_K blk.8.attn_v.weight q5_K -blk.9.ffn_down.weight q5_K +blk.8.ffn_down.weight q5_K blk.9.attn_output.weight q5_K blk.9.attn_v.weight q5_K +blk.9.ffn_down.weight q5_K +blk.10.attn_output.weight q5_K +blk.10.attn_v.weight q5_K +blk.10.ffn_down.weight q5_K +blk.11.attn_output.weight q5_K +blk.11.attn_v.weight q5_K +blk.11.ffn_down.weight q5_K +blk.12.attn_output.weight q5_K +blk.12.attn_v.weight q5_K +blk.12.ffn_down.weight q5_K +blk.13.attn_output.weight q5_K +blk.13.attn_v.weight q5_K +blk.13.ffn_down.weight q5_K +blk.14.attn_output.weight q5_K +blk.14.attn_v.weight q5_K blk.14.ffn_down.weight q5_K -blk.15.ffn_down.weight q5_K blk.15.attn_output.weight q5_K blk.15.attn_v.weight q5_K -blk.16.ffn_down.weight q5_K +blk.15.ffn_down.weight q5_K blk.16.attn_output.weight q5_K blk.16.attn_v.weight q5_K -blk.17.ffn_down.weight q5_K +blk.16.ffn_down.weight q5_K blk.17.attn_output.weight q5_K blk.17.attn_v.weight q5_K -blk.18.ffn_down.weight q5_K +blk.17.ffn_down.weight q5_K blk.18.attn_output.weight q5_K blk.18.attn_v.weight q5_K -blk.19.ffn_down.weight q5_K +blk.18.ffn_down.weight q5_K blk.19.attn_output.weight q5_K blk.19.attn_v.weight q5_K -blk.20.ffn_down.weight q5_K +blk.19.ffn_down.weight q5_K blk.20.attn_output.weight q5_K blk.20.attn_v.weight q5_K -blk.21.ffn_down.weight q5_K +blk.20.ffn_down.weight q5_K blk.21.attn_output.weight q5_K blk.21.attn_v.weight q5_K -blk.22.ffn_down.weight q5_K +blk.21.ffn_down.weight q5_K blk.22.attn_output.weight q5_K blk.22.attn_v.weight q5_K -blk.23.ffn_down.weight q5_K +blk.22.ffn_down.weight q5_K blk.23.attn_output.weight q5_K blk.23.attn_v.weight q5_K -blk.24.ffn_down.weight q5_K +blk.23.ffn_down.weight q5_K blk.24.attn_output.weight q5_K blk.24.attn_v.weight q5_K -blk.25.ffn_down.weight q5_K +blk.24.ffn_down.weight q5_K blk.25.attn_output.weight q5_K blk.25.attn_v.weight q5_K -blk.26.ffn_down.weight q5_K +blk.25.ffn_down.weight q5_K blk.26.attn_output.weight q5_K blk.26.attn_v.weight q5_K -blk.27.ffn_down.weight q5_K +blk.26.ffn_down.weight q5_K blk.27.attn_output.weight q5_K blk.27.attn_v.weight q5_K -blk.28.ffn_down.weight q5_K +blk.27.ffn_down.weight q5_K blk.28.attn_output.weight q5_K blk.28.attn_v.weight q5_K -blk.29.ffn_down.weight q5_K +blk.28.ffn_down.weight q5_K blk.29.attn_output.weight q5_K blk.29.attn_v.weight q5_K -blk.30.ffn_down.weight q5_K +blk.29.ffn_down.weight q5_K blk.30.attn_output.weight q5_K blk.30.attn_v.weight q5_K -blk.31.ffn_down.weight q5_K +blk.30.ffn_down.weight q5_K blk.31.attn_output.weight q5_K blk.31.attn_v.weight q5_K -blk.32.ffn_down.weight q5_K +blk.31.ffn_down.weight q5_K blk.32.attn_output.weight q5_K blk.32.attn_v.weight q5_K -blk.33.ffn_down.weight q5_K +blk.32.ffn_down.weight q5_K blk.33.attn_output.weight q5_K blk.33.attn_v.weight q5_K +blk.33.ffn_down.weight q5_K [Q4_K_S] q4_K token_embd.weight q6_K -blk.0.ffn_down.weight q5_K blk.0.attn_v.weight q5_K -blk.1.ffn_down.weight q5_K +blk.0.ffn_down.weight q5_K blk.1.attn_v.weight q5_K -blk.10.ffn_down.weight q5_K -blk.10.attn_v.weight q5_K -blk.11.ffn_down.weight q5_K -blk.11.attn_v.weight q5_K +blk.1.ffn_down.weight q5_K +blk.2.attn_v.weight q5_K +blk.2.ffn_down.weight q5_K +blk.3.attn_v.weight q5_K +blk.3.ffn_down.weight q5_K [Q4_K_M] q4_K token_embd.weight q6_K -blk.0.ffn_down.weight q6_K blk.0.attn_v.weight q6_K -blk.1.ffn_down.weight q6_K +blk.0.ffn_down.weight q6_K blk.1.attn_v.weight q6_K -blk.10.ffn_down.weight q6_K -blk.10.attn_v.weight q6_K -blk.11.ffn_down.weight q6_K -blk.11.attn_v.weight q6_K -blk.14.attn_v.weight q6_K +blk.1.ffn_down.weight q6_K +blk.2.attn_v.weight q6_K blk.2.ffn_down.weight q6_K -blk.4.attn_v.weight q6_K -blk.5.ffn_down.weight q6_K -blk.7.attn_v.weight q6_K -blk.8.ffn_down.weight q6_K -blk.15.ffn_down.weight q6_K +blk.3.attn_v.weight q6_K +blk.3.ffn_down.weight q6_K +blk.6.attn_v.weight q6_K +blk.6.ffn_down.weight q6_K +blk.9.attn_v.weight q6_K +blk.9.ffn_down.weight q6_K +blk.12.attn_v.weight q6_K +blk.12.ffn_down.weight q6_K blk.15.attn_v.weight q6_K -blk.18.ffn_down.weight q6_K +blk.15.ffn_down.weight q6_K blk.18.attn_v.weight q6_K -blk.21.ffn_down.weight q6_K +blk.18.ffn_down.weight q6_K blk.21.attn_v.weight q6_K -blk.24.ffn_down.weight q6_K +blk.21.ffn_down.weight q6_K blk.24.attn_v.weight q6_K -blk.27.ffn_down.weight q6_K +blk.24.ffn_down.weight q6_K blk.27.attn_v.weight q6_K -blk.29.ffn_down.weight q6_K +blk.27.ffn_down.weight q6_K blk.29.attn_v.weight q6_K -blk.30.ffn_down.weight q6_K +blk.29.ffn_down.weight q6_K blk.30.attn_v.weight q6_K -blk.31.ffn_down.weight q6_K +blk.30.ffn_down.weight q6_K blk.31.attn_v.weight q6_K -blk.32.ffn_down.weight q6_K +blk.31.ffn_down.weight q6_K blk.32.attn_v.weight q6_K -blk.33.ffn_down.weight q6_K +blk.32.ffn_down.weight q6_K blk.33.attn_v.weight q6_K +blk.33.ffn_down.weight q6_K [Q5_K_S] q5_K token_embd.weight q6_K [Q5_K_M] q5_K token_embd.weight q6_K -blk.0.ffn_down.weight q6_K blk.0.attn_v.weight q6_K -blk.1.ffn_down.weight q6_K +blk.0.ffn_down.weight q6_K blk.1.attn_v.weight q6_K -blk.10.ffn_down.weight q6_K -blk.10.attn_v.weight q6_K -blk.11.ffn_down.weight q6_K -blk.11.attn_v.weight q6_K -blk.14.attn_v.weight q6_K +blk.1.ffn_down.weight q6_K +blk.2.attn_v.weight q6_K blk.2.ffn_down.weight q6_K -blk.4.attn_v.weight q6_K -blk.5.ffn_down.weight q6_K -blk.7.attn_v.weight q6_K -blk.8.ffn_down.weight q6_K -blk.15.ffn_down.weight q6_K +blk.3.attn_v.weight q6_K +blk.3.ffn_down.weight q6_K +blk.6.attn_v.weight q6_K +blk.6.ffn_down.weight q6_K +blk.9.attn_v.weight q6_K +blk.9.ffn_down.weight q6_K +blk.12.attn_v.weight q6_K +blk.12.ffn_down.weight q6_K blk.15.attn_v.weight q6_K -blk.18.ffn_down.weight q6_K +blk.15.ffn_down.weight q6_K blk.18.attn_v.weight q6_K -blk.21.ffn_down.weight q6_K +blk.18.ffn_down.weight q6_K blk.21.attn_v.weight q6_K -blk.24.ffn_down.weight q6_K +blk.21.ffn_down.weight q6_K blk.24.attn_v.weight q6_K -blk.27.ffn_down.weight q6_K +blk.24.ffn_down.weight q6_K blk.27.attn_v.weight q6_K -blk.29.ffn_down.weight q6_K +blk.27.ffn_down.weight q6_K blk.29.attn_v.weight q6_K -blk.30.ffn_down.weight q6_K +blk.29.ffn_down.weight q6_K blk.30.attn_v.weight q6_K -blk.31.ffn_down.weight q6_K +blk.30.ffn_down.weight q6_K blk.31.attn_v.weight q6_K -blk.32.ffn_down.weight q6_K +blk.31.ffn_down.weight q6_K blk.32.attn_v.weight q6_K -blk.33.ffn_down.weight q6_K +blk.32.ffn_down.weight q6_K blk.33.attn_v.weight q6_K +blk.33.ffn_down.weight q6_K [Q6_K] q6_K [IQ2_XXS] iq2_xxs token_embd.weight q5_K -blk.0.ffn_down.weight q2_K blk.0.attn_v.weight q2_K -blk.1.ffn_down.weight q2_K +blk.0.ffn_down.weight q2_K blk.1.attn_v.weight q2_K -blk.10.ffn_down.weight q2_K -blk.10.attn_v.weight q2_K -blk.11.ffn_down.weight q2_K -blk.11.attn_v.weight q2_K -blk.12.attn_v.weight q2_K -blk.13.attn_v.weight q2_K -blk.14.attn_v.weight q2_K +blk.1.ffn_down.weight q2_K blk.2.attn_v.weight q2_K +blk.2.ffn_down.weight q2_K blk.3.attn_v.weight q2_K +blk.3.ffn_down.weight q2_K blk.4.attn_v.weight q2_K blk.5.attn_v.weight q2_K blk.6.attn_v.weight q2_K blk.7.attn_v.weight q2_K blk.8.attn_v.weight q2_K blk.9.attn_v.weight q2_K +blk.10.attn_v.weight q2_K +blk.11.attn_v.weight q2_K +blk.12.attn_v.weight q2_K +blk.13.attn_v.weight q2_K +blk.14.attn_v.weight q2_K blk.15.attn_v.weight q2_K blk.16.attn_v.weight q2_K blk.17.attn_v.weight q2_K @@ -470,25 +470,25 @@ blk.33.attn_v.weight q2_K [IQ2_XS] iq2_xs token_embd.weight q5_K -blk.0.ffn_down.weight q2_K blk.0.attn_v.weight q2_K -blk.1.ffn_down.weight q2_K +blk.0.ffn_down.weight q2_K blk.1.attn_v.weight q2_K -blk.10.ffn_down.weight q2_K -blk.10.attn_v.weight q2_K -blk.11.ffn_down.weight q2_K -blk.11.attn_v.weight q2_K -blk.12.attn_v.weight q2_K -blk.13.attn_v.weight q2_K -blk.14.attn_v.weight q2_K +blk.1.ffn_down.weight q2_K blk.2.attn_v.weight q2_K +blk.2.ffn_down.weight q2_K blk.3.attn_v.weight q2_K +blk.3.ffn_down.weight q2_K blk.4.attn_v.weight q2_K blk.5.attn_v.weight q2_K blk.6.attn_v.weight q2_K blk.7.attn_v.weight q2_K blk.8.attn_v.weight q2_K blk.9.attn_v.weight q2_K +blk.10.attn_v.weight q2_K +blk.11.attn_v.weight q2_K +blk.12.attn_v.weight q2_K +blk.13.attn_v.weight q2_K +blk.14.attn_v.weight q2_K blk.15.attn_v.weight q2_K blk.16.attn_v.weight q2_K blk.17.attn_v.weight q2_K @@ -513,8 +513,8 @@ blk.33.attn_v.weight q2_K token_embd.weight q6_K blk.0.ffn_down.weight q4_K blk.1.ffn_down.weight q4_K -blk.10.ffn_down.weight q4_K -blk.11.ffn_down.weight q4_K +blk.2.ffn_down.weight q4_K +blk.3.ffn_down.weight q4_K [IQ3_XS] iq3_s token_embd.weight q6_K @@ -522,110 +522,110 @@ blk.0.attn_k.weight iq3_xxs blk.0.attn_q.weight iq3_xxs blk.1.attn_k.weight iq3_xxs blk.1.attn_q.weight iq3_xxs -blk.10.attn_k.weight iq3_xxs -blk.10.attn_q.weight iq3_xxs -blk.11.attn_k.weight iq3_xxs -blk.11.attn_q.weight iq3_xxs -blk.12.ffn_gate.weight iq3_xxs -blk.12.ffn_up.weight iq3_xxs -blk.12.attn_k.weight iq3_xxs -blk.12.attn_q.weight iq3_xxs -blk.13.ffn_gate.weight iq3_xxs -blk.13.ffn_up.weight iq3_xxs -blk.13.attn_k.weight iq3_xxs -blk.13.attn_q.weight iq3_xxs -blk.14.ffn_gate.weight iq3_xxs -blk.14.ffn_up.weight iq3_xxs -blk.14.attn_k.weight iq3_xxs -blk.14.attn_q.weight iq3_xxs -blk.2.ffn_gate.weight iq3_xxs -blk.2.ffn_up.weight iq3_xxs blk.2.attn_k.weight iq3_xxs blk.2.attn_q.weight iq3_xxs -blk.3.ffn_gate.weight iq3_xxs -blk.3.ffn_up.weight iq3_xxs blk.3.attn_k.weight iq3_xxs blk.3.attn_q.weight iq3_xxs -blk.4.ffn_gate.weight iq3_xxs -blk.4.ffn_up.weight iq3_xxs blk.4.attn_k.weight iq3_xxs blk.4.attn_q.weight iq3_xxs -blk.5.ffn_gate.weight iq3_xxs -blk.5.ffn_up.weight iq3_xxs +blk.4.ffn_gate.weight iq3_xxs +blk.4.ffn_up.weight iq3_xxs blk.5.attn_k.weight iq3_xxs blk.5.attn_q.weight iq3_xxs -blk.6.ffn_gate.weight iq3_xxs -blk.6.ffn_up.weight iq3_xxs +blk.5.ffn_gate.weight iq3_xxs +blk.5.ffn_up.weight iq3_xxs blk.6.attn_k.weight iq3_xxs blk.6.attn_q.weight iq3_xxs -blk.7.ffn_gate.weight iq3_xxs -blk.7.ffn_up.weight iq3_xxs +blk.6.ffn_gate.weight iq3_xxs +blk.6.ffn_up.weight iq3_xxs blk.7.attn_k.weight iq3_xxs blk.7.attn_q.weight iq3_xxs -blk.8.ffn_gate.weight iq3_xxs -blk.8.ffn_up.weight iq3_xxs +blk.7.ffn_gate.weight iq3_xxs +blk.7.ffn_up.weight iq3_xxs blk.8.attn_k.weight iq3_xxs blk.8.attn_q.weight iq3_xxs -blk.9.ffn_gate.weight iq3_xxs -blk.9.ffn_up.weight iq3_xxs +blk.8.ffn_gate.weight iq3_xxs +blk.8.ffn_up.weight iq3_xxs blk.9.attn_k.weight iq3_xxs blk.9.attn_q.weight iq3_xxs -blk.15.ffn_gate.weight iq3_xxs -blk.15.ffn_up.weight iq3_xxs +blk.9.ffn_gate.weight iq3_xxs +blk.9.ffn_up.weight iq3_xxs +blk.10.attn_k.weight iq3_xxs +blk.10.attn_q.weight iq3_xxs +blk.10.ffn_gate.weight iq3_xxs +blk.10.ffn_up.weight iq3_xxs +blk.11.attn_k.weight iq3_xxs +blk.11.attn_q.weight iq3_xxs +blk.11.ffn_gate.weight iq3_xxs +blk.11.ffn_up.weight iq3_xxs +blk.12.attn_k.weight iq3_xxs +blk.12.attn_q.weight iq3_xxs +blk.12.ffn_gate.weight iq3_xxs +blk.12.ffn_up.weight iq3_xxs +blk.13.attn_k.weight iq3_xxs +blk.13.attn_q.weight iq3_xxs +blk.13.ffn_gate.weight iq3_xxs +blk.13.ffn_up.weight iq3_xxs +blk.14.attn_k.weight iq3_xxs +blk.14.attn_q.weight iq3_xxs +blk.14.ffn_gate.weight iq3_xxs +blk.14.ffn_up.weight iq3_xxs blk.15.attn_k.weight iq3_xxs blk.15.attn_q.weight iq3_xxs -blk.16.ffn_gate.weight iq3_xxs -blk.16.ffn_up.weight iq3_xxs +blk.15.ffn_gate.weight iq3_xxs +blk.15.ffn_up.weight iq3_xxs blk.16.attn_k.weight iq3_xxs blk.16.attn_q.weight iq3_xxs -blk.17.ffn_gate.weight iq3_xxs -blk.17.ffn_up.weight iq3_xxs +blk.16.ffn_gate.weight iq3_xxs +blk.16.ffn_up.weight iq3_xxs blk.17.attn_k.weight iq3_xxs blk.17.attn_q.weight iq3_xxs -blk.18.ffn_gate.weight iq3_xxs -blk.18.ffn_up.weight iq3_xxs +blk.17.ffn_gate.weight iq3_xxs +blk.17.ffn_up.weight iq3_xxs blk.18.attn_k.weight iq3_xxs blk.18.attn_q.weight iq3_xxs -blk.19.ffn_gate.weight iq3_xxs -blk.19.ffn_up.weight iq3_xxs +blk.18.ffn_gate.weight iq3_xxs +blk.18.ffn_up.weight iq3_xxs blk.19.attn_k.weight iq3_xxs blk.19.attn_q.weight iq3_xxs -blk.20.ffn_gate.weight iq3_xxs -blk.20.ffn_up.weight iq3_xxs +blk.19.ffn_gate.weight iq3_xxs +blk.19.ffn_up.weight iq3_xxs blk.20.attn_k.weight iq3_xxs blk.20.attn_q.weight iq3_xxs -blk.21.ffn_gate.weight iq3_xxs -blk.21.ffn_up.weight iq3_xxs +blk.20.ffn_gate.weight iq3_xxs +blk.20.ffn_up.weight iq3_xxs blk.21.attn_k.weight iq3_xxs blk.21.attn_q.weight iq3_xxs -blk.22.ffn_gate.weight iq3_xxs -blk.22.ffn_up.weight iq3_xxs +blk.21.ffn_gate.weight iq3_xxs +blk.21.ffn_up.weight iq3_xxs blk.22.attn_k.weight iq3_xxs blk.22.attn_q.weight iq3_xxs -blk.23.ffn_gate.weight iq3_xxs -blk.23.ffn_up.weight iq3_xxs +blk.22.ffn_gate.weight iq3_xxs +blk.22.ffn_up.weight iq3_xxs blk.23.attn_k.weight iq3_xxs blk.23.attn_q.weight iq3_xxs -blk.24.ffn_gate.weight iq3_xxs -blk.24.ffn_up.weight iq3_xxs +blk.23.ffn_gate.weight iq3_xxs +blk.23.ffn_up.weight iq3_xxs blk.24.attn_k.weight iq3_xxs blk.24.attn_q.weight iq3_xxs -blk.25.ffn_gate.weight iq3_xxs -blk.25.ffn_up.weight iq3_xxs +blk.24.ffn_gate.weight iq3_xxs +blk.24.ffn_up.weight iq3_xxs blk.25.attn_k.weight iq3_xxs blk.25.attn_q.weight iq3_xxs -blk.26.ffn_gate.weight iq3_xxs -blk.26.ffn_up.weight iq3_xxs +blk.25.ffn_gate.weight iq3_xxs +blk.25.ffn_up.weight iq3_xxs blk.26.attn_k.weight iq3_xxs blk.26.attn_q.weight iq3_xxs -blk.27.ffn_gate.weight iq3_xxs -blk.27.ffn_up.weight iq3_xxs +blk.26.ffn_gate.weight iq3_xxs +blk.26.ffn_up.weight iq3_xxs blk.27.attn_k.weight iq3_xxs blk.27.attn_q.weight iq3_xxs -blk.28.ffn_gate.weight iq3_xxs -blk.28.ffn_up.weight iq3_xxs +blk.27.ffn_gate.weight iq3_xxs +blk.27.ffn_up.weight iq3_xxs blk.28.attn_k.weight iq3_xxs blk.28.attn_q.weight iq3_xxs +blk.28.ffn_gate.weight iq3_xxs +blk.28.ffn_up.weight iq3_xxs blk.29.attn_k.weight iq3_xxs blk.29.attn_q.weight iq3_xxs blk.30.attn_k.weight iq3_xxs @@ -639,201 +639,191 @@ blk.33.attn_q.weight iq3_xxs [IQ3_XXS] iq3_xxs token_embd.weight q5_K -blk.0.ffn_down.weight q4_K blk.0.attn_k.weight iq2_s blk.0.attn_output.weight iq3_s blk.0.attn_q.weight iq2_s blk.0.attn_v.weight iq3_s -blk.1.ffn_down.weight q4_K +blk.0.ffn_down.weight q4_K blk.1.attn_k.weight iq2_s blk.1.attn_output.weight iq3_s blk.1.attn_q.weight iq2_s blk.1.attn_v.weight iq3_s -blk.10.ffn_down.weight q4_K -blk.10.attn_k.weight iq2_s -blk.10.attn_output.weight iq3_s -blk.10.attn_q.weight iq2_s -blk.10.attn_v.weight iq3_s -blk.11.ffn_down.weight q4_K -blk.11.attn_k.weight iq2_s -blk.11.attn_output.weight iq3_s -blk.11.attn_q.weight iq2_s -blk.11.attn_v.weight iq3_s -blk.12.ffn_down.weight q3_K -blk.12.attn_k.weight iq2_s -blk.12.attn_output.weight iq3_s -blk.12.attn_q.weight iq2_s -blk.12.attn_v.weight iq3_s -blk.13.ffn_down.weight q3_K -blk.13.attn_k.weight iq2_s -blk.13.attn_output.weight iq3_s -blk.13.attn_q.weight iq2_s -blk.13.attn_v.weight iq3_s -blk.14.attn_k.weight iq2_s -blk.14.attn_output.weight iq3_s -blk.14.attn_q.weight iq2_s -blk.14.attn_v.weight iq3_s -blk.2.ffn_down.weight q3_K +blk.1.ffn_down.weight q4_K blk.2.attn_k.weight iq2_s blk.2.attn_output.weight iq3_s blk.2.attn_q.weight iq2_s blk.2.attn_v.weight iq3_s -blk.3.ffn_down.weight q3_K +blk.2.ffn_down.weight q4_K blk.3.attn_k.weight iq2_s blk.3.attn_output.weight iq3_s blk.3.attn_q.weight iq2_s blk.3.attn_v.weight iq3_s -blk.4.ffn_down.weight q3_K +blk.3.ffn_down.weight q4_K blk.4.attn_k.weight iq2_s blk.4.attn_output.weight iq3_s blk.4.attn_q.weight iq2_s blk.4.attn_v.weight iq3_s -blk.5.ffn_down.weight q3_K +blk.4.ffn_down.weight q3_K blk.5.attn_k.weight iq2_s blk.5.attn_output.weight iq3_s blk.5.attn_q.weight iq2_s blk.5.attn_v.weight iq3_s -blk.6.ffn_down.weight q3_K +blk.5.ffn_down.weight q3_K blk.6.attn_k.weight iq2_s blk.6.attn_output.weight iq3_s blk.6.attn_q.weight iq2_s blk.6.attn_v.weight iq3_s -blk.7.ffn_down.weight q3_K +blk.6.ffn_down.weight q3_K blk.7.attn_k.weight iq2_s blk.7.attn_output.weight iq3_s blk.7.attn_q.weight iq2_s blk.7.attn_v.weight iq3_s -blk.8.ffn_down.weight q3_K +blk.7.ffn_down.weight q3_K blk.8.attn_k.weight iq2_s blk.8.attn_output.weight iq3_s blk.8.attn_q.weight iq2_s blk.8.attn_v.weight iq3_s -blk.9.ffn_down.weight q3_K +blk.8.ffn_down.weight q3_K blk.9.attn_k.weight iq2_s blk.9.attn_output.weight iq3_s blk.9.attn_q.weight iq2_s blk.9.attn_v.weight iq3_s +blk.9.ffn_down.weight q3_K +blk.10.attn_k.weight iq2_s +blk.10.attn_output.weight iq3_s +blk.10.attn_q.weight iq2_s +blk.10.attn_v.weight iq3_s +blk.10.ffn_down.weight q3_K +blk.11.attn_k.weight iq2_s +blk.11.attn_output.weight iq3_s +blk.11.attn_q.weight iq2_s +blk.11.attn_v.weight iq3_s +blk.11.ffn_down.weight q3_K +blk.12.attn_k.weight iq2_s +blk.12.attn_output.weight iq3_s +blk.12.attn_q.weight iq2_s +blk.12.attn_v.weight iq3_s +blk.12.ffn_down.weight q3_K +blk.13.attn_k.weight iq2_s +blk.13.attn_output.weight iq3_s +blk.13.attn_q.weight iq2_s +blk.13.attn_v.weight iq3_s +blk.13.ffn_down.weight q3_K +blk.14.attn_k.weight iq2_s +blk.14.attn_output.weight iq3_s +blk.14.attn_q.weight iq2_s +blk.14.attn_v.weight iq3_s blk.14.ffn_down.weight q3_K -blk.15.ffn_down.weight q3_K blk.15.attn_k.weight iq2_s blk.15.attn_output.weight iq3_s blk.15.attn_q.weight iq2_s blk.15.attn_v.weight iq3_s -blk.16.ffn_down.weight q3_K +blk.15.ffn_down.weight q3_K blk.16.attn_k.weight iq2_s blk.16.attn_output.weight iq3_s blk.16.attn_q.weight iq2_s blk.16.attn_v.weight iq3_s -blk.17.ffn_down.weight q3_K +blk.16.ffn_down.weight q3_K blk.17.attn_k.weight iq2_s blk.17.attn_output.weight iq3_s blk.17.attn_q.weight iq2_s blk.17.attn_v.weight iq3_s -blk.18.ffn_down.weight q3_K +blk.17.ffn_down.weight q3_K blk.18.attn_k.weight iq2_s blk.18.attn_output.weight iq3_s blk.18.attn_q.weight iq2_s blk.18.attn_v.weight iq3_s -blk.19.ffn_down.weight q3_K +blk.18.ffn_down.weight q3_K blk.19.attn_k.weight iq2_s blk.19.attn_output.weight iq3_s blk.19.attn_q.weight iq2_s blk.19.attn_v.weight iq3_s -blk.20.ffn_down.weight q3_K +blk.19.ffn_down.weight q3_K blk.20.attn_k.weight iq2_s blk.20.attn_output.weight iq3_s blk.20.attn_q.weight iq2_s blk.20.attn_v.weight iq3_s -blk.21.ffn_down.weight q3_K +blk.20.ffn_down.weight q3_K blk.21.attn_k.weight iq2_s blk.21.attn_output.weight iq3_s blk.21.attn_q.weight iq2_s blk.21.attn_v.weight iq3_s -blk.22.ffn_down.weight q3_K +blk.21.ffn_down.weight q3_K blk.22.attn_k.weight iq2_s blk.22.attn_output.weight iq3_s blk.22.attn_q.weight iq2_s blk.22.attn_v.weight iq3_s -blk.23.ffn_down.weight q3_K +blk.22.ffn_down.weight q3_K blk.23.attn_k.weight iq2_s blk.23.attn_output.weight iq3_s blk.23.attn_q.weight iq2_s blk.23.attn_v.weight iq3_s -blk.24.ffn_down.weight q3_K +blk.23.ffn_down.weight q3_K blk.24.attn_k.weight iq2_s blk.24.attn_output.weight iq3_s blk.24.attn_q.weight iq2_s blk.24.attn_v.weight iq3_s -blk.25.ffn_down.weight q3_K +blk.24.ffn_down.weight q3_K blk.25.attn_k.weight iq2_s blk.25.attn_output.weight iq3_s blk.25.attn_q.weight iq2_s blk.25.attn_v.weight iq3_s -blk.26.ffn_down.weight q3_K +blk.25.ffn_down.weight q3_K blk.26.attn_k.weight iq2_s blk.26.attn_output.weight iq3_s blk.26.attn_q.weight iq2_s blk.26.attn_v.weight iq3_s -blk.27.ffn_down.weight q3_K +blk.26.ffn_down.weight q3_K blk.27.attn_k.weight iq2_s blk.27.attn_output.weight iq3_s blk.27.attn_q.weight iq2_s blk.27.attn_v.weight iq3_s -blk.28.ffn_down.weight q3_K +blk.27.ffn_down.weight q3_K blk.28.attn_k.weight iq2_s blk.28.attn_output.weight iq3_s blk.28.attn_q.weight iq2_s blk.28.attn_v.weight iq3_s -blk.29.ffn_down.weight q3_K +blk.28.ffn_down.weight q3_K blk.29.attn_k.weight iq2_s blk.29.attn_output.weight iq3_s blk.29.attn_q.weight iq2_s blk.29.attn_v.weight iq3_s -blk.30.ffn_down.weight q3_K +blk.29.ffn_down.weight q3_K blk.30.attn_k.weight iq2_s blk.30.attn_output.weight iq3_s blk.30.attn_q.weight iq2_s blk.30.attn_v.weight iq3_s -blk.31.ffn_down.weight q3_K +blk.30.ffn_down.weight q3_K blk.31.attn_k.weight iq2_s blk.31.attn_output.weight iq3_s blk.31.attn_q.weight iq2_s blk.31.attn_v.weight iq3_s -blk.32.ffn_down.weight q3_K +blk.31.ffn_down.weight q3_K blk.32.attn_k.weight iq2_s blk.32.attn_output.weight iq3_s blk.32.attn_q.weight iq2_s blk.32.attn_v.weight iq3_s -blk.33.ffn_down.weight q3_K +blk.32.ffn_down.weight q3_K blk.33.attn_k.weight iq2_s blk.33.attn_output.weight iq3_s blk.33.attn_q.weight iq2_s blk.33.attn_v.weight iq3_s +blk.33.ffn_down.weight q3_K [IQ1_S] iq1_s token_embd.weight q5_K -blk.0.ffn_down.weight q2_K blk.0.attn_output.weight iq2_xxs blk.0.attn_v.weight q2_K -blk.1.ffn_down.weight q2_K +blk.0.ffn_down.weight q2_K blk.1.attn_output.weight iq2_xxs blk.1.attn_v.weight q2_K -blk.10.ffn_down.weight q2_K -blk.10.attn_output.weight iq2_xxs -blk.10.attn_v.weight q2_K -blk.11.ffn_down.weight q2_K -blk.11.attn_output.weight iq2_xxs -blk.11.attn_v.weight q2_K -blk.12.attn_output.weight iq2_xxs -blk.12.attn_v.weight q2_K -blk.13.attn_output.weight iq2_xxs -blk.13.attn_v.weight q2_K -blk.14.attn_output.weight iq2_xxs -blk.14.attn_v.weight q2_K +blk.1.ffn_down.weight q2_K blk.2.attn_output.weight iq2_xxs blk.2.attn_v.weight q2_K +blk.2.ffn_down.weight q2_K blk.3.attn_output.weight iq2_xxs blk.3.attn_v.weight q2_K +blk.3.ffn_down.weight q2_K blk.4.attn_output.weight iq2_xxs blk.4.attn_v.weight q2_K blk.5.attn_output.weight iq2_xxs @@ -846,6 +836,16 @@ blk.8.attn_output.weight iq2_xxs blk.8.attn_v.weight q2_K blk.9.attn_output.weight iq2_xxs blk.9.attn_v.weight q2_K +blk.10.attn_output.weight iq2_xxs +blk.10.attn_v.weight q2_K +blk.11.attn_output.weight iq2_xxs +blk.11.attn_v.weight q2_K +blk.12.attn_output.weight iq2_xxs +blk.12.attn_v.weight q2_K +blk.13.attn_output.weight iq2_xxs +blk.13.attn_v.weight q2_K +blk.14.attn_output.weight iq2_xxs +blk.14.attn_v.weight q2_K blk.15.attn_output.weight iq2_xxs blk.15.attn_v.weight q2_K blk.16.attn_output.weight iq2_xxs @@ -889,36 +889,26 @@ blk.33.attn_v.weight q2_K token_embd.weight q6_K blk.0.ffn_down.weight q5_K blk.1.ffn_down.weight q5_K -blk.10.ffn_down.weight q5_K -blk.11.ffn_down.weight q5_K +blk.2.ffn_down.weight q5_K +blk.3.ffn_down.weight q5_K [IQ3_S] iq3_s token_embd.weight q6_K [IQ3_M] iq3_s token_embd.weight q6_K -blk.0.ffn_down.weight q4_K blk.0.attn_output.weight q4_K blk.0.attn_v.weight q4_K -blk.1.ffn_down.weight q4_K +blk.0.ffn_down.weight q4_K blk.1.attn_output.weight q4_K blk.1.attn_v.weight q4_K -blk.10.ffn_down.weight q4_K -blk.10.attn_output.weight q4_K -blk.10.attn_v.weight q4_K -blk.11.ffn_down.weight q4_K -blk.11.attn_output.weight q4_K -blk.11.attn_v.weight q4_K -blk.12.attn_output.weight q4_K -blk.12.attn_v.weight q4_K -blk.13.attn_output.weight q4_K -blk.13.attn_v.weight q4_K -blk.14.attn_output.weight q4_K -blk.14.attn_v.weight q4_K +blk.1.ffn_down.weight q4_K blk.2.attn_output.weight q4_K blk.2.attn_v.weight q4_K +blk.2.ffn_down.weight q4_K blk.3.attn_output.weight q4_K blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight q4_K blk.4.attn_output.weight q4_K blk.4.attn_v.weight q4_K blk.5.attn_output.weight q4_K @@ -931,6 +921,16 @@ blk.8.attn_output.weight q4_K blk.8.attn_v.weight q4_K blk.9.attn_output.weight q4_K blk.9.attn_v.weight q4_K +blk.10.attn_output.weight q4_K +blk.10.attn_v.weight q4_K +blk.11.attn_output.weight q4_K +blk.11.attn_v.weight q4_K +blk.12.attn_output.weight q4_K +blk.12.attn_v.weight q4_K +blk.13.attn_output.weight q4_K +blk.13.attn_v.weight q4_K +blk.14.attn_output.weight q4_K +blk.14.attn_v.weight q4_K blk.15.attn_output.weight q4_K blk.15.attn_v.weight q4_K blk.16.attn_output.weight q4_K @@ -972,28 +972,18 @@ blk.33.attn_v.weight q4_K [IQ2_S] iq2_xs token_embd.weight q5_K -blk.0.ffn_down.weight iq3_s blk.0.attn_output.weight iq3_s blk.0.attn_v.weight iq3_s -blk.1.ffn_down.weight iq3_s +blk.0.ffn_down.weight iq3_s blk.1.attn_output.weight iq3_s blk.1.attn_v.weight iq3_s -blk.10.ffn_down.weight iq3_s -blk.10.attn_output.weight iq3_s -blk.10.attn_v.weight iq3_s -blk.11.ffn_down.weight iq3_s -blk.11.attn_output.weight iq3_s -blk.11.attn_v.weight iq3_s -blk.12.attn_output.weight iq3_s -blk.12.attn_v.weight iq3_s -blk.13.attn_output.weight iq3_s -blk.13.attn_v.weight iq3_s -blk.14.attn_output.weight iq3_s -blk.14.attn_v.weight iq3_s +blk.1.ffn_down.weight iq3_s blk.2.attn_output.weight iq3_s blk.2.attn_v.weight iq3_s +blk.2.ffn_down.weight iq3_s blk.3.attn_output.weight iq3_s blk.3.attn_v.weight iq3_s +blk.3.ffn_down.weight iq3_s blk.4.attn_output.weight iq3_s blk.4.attn_v.weight iq3_s blk.5.attn_output.weight iq3_s @@ -1006,6 +996,16 @@ blk.8.attn_output.weight iq3_s blk.8.attn_v.weight iq3_s blk.9.attn_output.weight iq3_s blk.9.attn_v.weight iq3_s +blk.10.attn_output.weight iq3_s +blk.10.attn_v.weight iq3_s +blk.11.attn_output.weight iq3_s +blk.11.attn_v.weight iq3_s +blk.12.attn_output.weight iq3_s +blk.12.attn_v.weight iq3_s +blk.13.attn_output.weight iq3_s +blk.13.attn_v.weight iq3_s +blk.14.attn_output.weight iq3_s +blk.14.attn_v.weight iq3_s blk.15.attn_output.weight iq3_s blk.15.attn_v.weight iq3_s blk.16.attn_output.weight iq3_s @@ -1047,28 +1047,18 @@ blk.33.attn_v.weight iq3_s [IQ2_M] iq2_s token_embd.weight q5_K -blk.0.ffn_down.weight iq3_s blk.0.attn_output.weight iq3_s blk.0.attn_v.weight iq3_s -blk.1.ffn_down.weight iq3_s +blk.0.ffn_down.weight iq3_s blk.1.attn_output.weight iq3_s blk.1.attn_v.weight iq3_s -blk.10.ffn_down.weight iq3_s -blk.10.attn_output.weight iq3_s -blk.10.attn_v.weight iq3_s -blk.11.ffn_down.weight iq3_s -blk.11.attn_output.weight iq3_s -blk.11.attn_v.weight iq3_s -blk.12.attn_output.weight iq3_s -blk.12.attn_v.weight iq3_s -blk.13.attn_output.weight iq3_s -blk.13.attn_v.weight iq3_s -blk.14.attn_output.weight iq3_s -blk.14.attn_v.weight iq3_s +blk.1.ffn_down.weight iq3_s blk.2.attn_output.weight iq3_s blk.2.attn_v.weight iq3_s +blk.2.ffn_down.weight iq3_s blk.3.attn_output.weight iq3_s blk.3.attn_v.weight iq3_s +blk.3.ffn_down.weight iq3_s blk.4.attn_output.weight iq3_s blk.4.attn_v.weight iq3_s blk.5.attn_output.weight iq3_s @@ -1081,6 +1071,16 @@ blk.8.attn_output.weight iq3_s blk.8.attn_v.weight iq3_s blk.9.attn_output.weight iq3_s blk.9.attn_v.weight iq3_s +blk.10.attn_output.weight iq3_s +blk.10.attn_v.weight iq3_s +blk.11.attn_output.weight iq3_s +blk.11.attn_v.weight iq3_s +blk.12.attn_output.weight iq3_s +blk.12.attn_v.weight iq3_s +blk.13.attn_output.weight iq3_s +blk.13.attn_v.weight iq3_s +blk.14.attn_output.weight iq3_s +blk.14.attn_v.weight iq3_s blk.15.attn_output.weight iq3_s blk.15.attn_v.weight iq3_s blk.16.attn_output.weight iq3_s @@ -1124,33 +1124,23 @@ blk.33.attn_v.weight iq3_s token_embd.weight q6_K blk.0.ffn_down.weight q5_K blk.1.ffn_down.weight q5_K -blk.10.ffn_down.weight q5_K -blk.11.ffn_down.weight q5_K +blk.2.ffn_down.weight q5_K +blk.3.ffn_down.weight q5_K [IQ1_M] iq1_m token_embd.weight q5_K -blk.0.ffn_down.weight q2_K blk.0.attn_output.weight iq2_xxs blk.0.attn_v.weight q2_K -blk.1.ffn_down.weight q2_K +blk.0.ffn_down.weight q2_K blk.1.attn_output.weight iq2_xxs blk.1.attn_v.weight q2_K -blk.10.ffn_down.weight q2_K -blk.10.attn_output.weight iq2_xxs -blk.10.attn_v.weight q2_K -blk.11.ffn_down.weight q2_K -blk.11.attn_output.weight iq2_xxs -blk.11.attn_v.weight q2_K -blk.12.attn_output.weight iq2_xxs -blk.12.attn_v.weight q2_K -blk.13.attn_output.weight iq2_xxs -blk.13.attn_v.weight q2_K -blk.14.attn_output.weight iq2_xxs -blk.14.attn_v.weight q2_K +blk.1.ffn_down.weight q2_K blk.2.attn_output.weight iq2_xxs blk.2.attn_v.weight q2_K +blk.2.ffn_down.weight q2_K blk.3.attn_output.weight iq2_xxs blk.3.attn_v.weight q2_K +blk.3.ffn_down.weight q2_K blk.4.attn_output.weight iq2_xxs blk.4.attn_v.weight q2_K blk.5.attn_output.weight iq2_xxs @@ -1163,6 +1153,16 @@ blk.8.attn_output.weight iq2_xxs blk.8.attn_v.weight q2_K blk.9.attn_output.weight iq2_xxs blk.9.attn_v.weight q2_K +blk.10.attn_output.weight iq2_xxs +blk.10.attn_v.weight q2_K +blk.11.attn_output.weight iq2_xxs +blk.11.attn_v.weight q2_K +blk.12.attn_output.weight iq2_xxs +blk.12.attn_v.weight q2_K +blk.13.attn_output.weight iq2_xxs +blk.13.attn_v.weight q2_K +blk.14.attn_output.weight iq2_xxs +blk.14.attn_v.weight q2_K blk.15.attn_output.weight iq2_xxs blk.15.attn_v.weight q2_K blk.16.attn_output.weight iq2_xxs @@ -1212,241 +1212,241 @@ token_embd.weight q6_K [MXFP4_MOE] mxfp4 token_embd.weight q8_0 -blk.0.ffn_down.weight q8_0 -blk.0.ffn_gate.weight q8_0 -blk.0.ffn_up.weight q8_0 blk.0.attn_k.weight q8_0 blk.0.attn_output.weight q8_0 blk.0.attn_q.weight q8_0 blk.0.attn_v.weight q8_0 -blk.1.ffn_down.weight q8_0 -blk.1.ffn_gate.weight q8_0 -blk.1.ffn_up.weight q8_0 +blk.0.ffn_down.weight q8_0 +blk.0.ffn_gate.weight q8_0 +blk.0.ffn_up.weight q8_0 blk.1.attn_k.weight q8_0 blk.1.attn_output.weight q8_0 blk.1.attn_q.weight q8_0 blk.1.attn_v.weight q8_0 -blk.10.ffn_down.weight q8_0 -blk.10.ffn_gate.weight q8_0 -blk.10.ffn_up.weight q8_0 -blk.10.attn_k.weight q8_0 -blk.10.attn_output.weight q8_0 -blk.10.attn_q.weight q8_0 -blk.10.attn_v.weight q8_0 -blk.11.ffn_down.weight q8_0 -blk.11.ffn_gate.weight q8_0 -blk.11.ffn_up.weight q8_0 -blk.11.attn_k.weight q8_0 -blk.11.attn_output.weight q8_0 -blk.11.attn_q.weight q8_0 -blk.11.attn_v.weight q8_0 -blk.12.ffn_down.weight q8_0 -blk.12.ffn_gate.weight q8_0 -blk.12.ffn_up.weight q8_0 -blk.12.attn_k.weight q8_0 -blk.12.attn_output.weight q8_0 -blk.12.attn_q.weight q8_0 -blk.12.attn_v.weight q8_0 -blk.13.ffn_down.weight q8_0 -blk.13.ffn_gate.weight q8_0 -blk.13.ffn_up.weight q8_0 -blk.13.attn_k.weight q8_0 -blk.13.attn_output.weight q8_0 -blk.13.attn_q.weight q8_0 -blk.13.attn_v.weight q8_0 -blk.14.ffn_gate.weight q8_0 -blk.14.ffn_up.weight q8_0 -blk.14.attn_k.weight q8_0 -blk.14.attn_output.weight q8_0 -blk.14.attn_q.weight q8_0 -blk.14.attn_v.weight q8_0 -blk.2.ffn_down.weight q8_0 -blk.2.ffn_gate.weight q8_0 -blk.2.ffn_up.weight q8_0 +blk.1.ffn_down.weight q8_0 +blk.1.ffn_gate.weight q8_0 +blk.1.ffn_up.weight q8_0 blk.2.attn_k.weight q8_0 blk.2.attn_output.weight q8_0 blk.2.attn_q.weight q8_0 blk.2.attn_v.weight q8_0 -blk.3.ffn_down.weight q8_0 -blk.3.ffn_gate.weight q8_0 -blk.3.ffn_up.weight q8_0 +blk.2.ffn_down.weight q8_0 +blk.2.ffn_gate.weight q8_0 +blk.2.ffn_up.weight q8_0 blk.3.attn_k.weight q8_0 blk.3.attn_output.weight q8_0 blk.3.attn_q.weight q8_0 blk.3.attn_v.weight q8_0 -blk.4.ffn_down.weight q8_0 -blk.4.ffn_gate.weight q8_0 -blk.4.ffn_up.weight q8_0 +blk.3.ffn_down.weight q8_0 +blk.3.ffn_gate.weight q8_0 +blk.3.ffn_up.weight q8_0 blk.4.attn_k.weight q8_0 blk.4.attn_output.weight q8_0 blk.4.attn_q.weight q8_0 blk.4.attn_v.weight q8_0 -blk.5.ffn_down.weight q8_0 -blk.5.ffn_gate.weight q8_0 -blk.5.ffn_up.weight q8_0 +blk.4.ffn_down.weight q8_0 +blk.4.ffn_gate.weight q8_0 +blk.4.ffn_up.weight q8_0 blk.5.attn_k.weight q8_0 blk.5.attn_output.weight q8_0 blk.5.attn_q.weight q8_0 blk.5.attn_v.weight q8_0 -blk.6.ffn_down.weight q8_0 -blk.6.ffn_gate.weight q8_0 -blk.6.ffn_up.weight q8_0 +blk.5.ffn_down.weight q8_0 +blk.5.ffn_gate.weight q8_0 +blk.5.ffn_up.weight q8_0 blk.6.attn_k.weight q8_0 blk.6.attn_output.weight q8_0 blk.6.attn_q.weight q8_0 blk.6.attn_v.weight q8_0 -blk.7.ffn_down.weight q8_0 -blk.7.ffn_gate.weight q8_0 -blk.7.ffn_up.weight q8_0 +blk.6.ffn_down.weight q8_0 +blk.6.ffn_gate.weight q8_0 +blk.6.ffn_up.weight q8_0 blk.7.attn_k.weight q8_0 blk.7.attn_output.weight q8_0 blk.7.attn_q.weight q8_0 blk.7.attn_v.weight q8_0 -blk.8.ffn_down.weight q8_0 -blk.8.ffn_gate.weight q8_0 -blk.8.ffn_up.weight q8_0 +blk.7.ffn_down.weight q8_0 +blk.7.ffn_gate.weight q8_0 +blk.7.ffn_up.weight q8_0 blk.8.attn_k.weight q8_0 blk.8.attn_output.weight q8_0 blk.8.attn_q.weight q8_0 blk.8.attn_v.weight q8_0 -blk.9.ffn_down.weight q8_0 -blk.9.ffn_gate.weight q8_0 -blk.9.ffn_up.weight q8_0 +blk.8.ffn_down.weight q8_0 +blk.8.ffn_gate.weight q8_0 +blk.8.ffn_up.weight q8_0 blk.9.attn_k.weight q8_0 blk.9.attn_output.weight q8_0 blk.9.attn_q.weight q8_0 blk.9.attn_v.weight q8_0 +blk.9.ffn_down.weight q8_0 +blk.9.ffn_gate.weight q8_0 +blk.9.ffn_up.weight q8_0 +blk.10.attn_k.weight q8_0 +blk.10.attn_output.weight q8_0 +blk.10.attn_q.weight q8_0 +blk.10.attn_v.weight q8_0 +blk.10.ffn_down.weight q8_0 +blk.10.ffn_gate.weight q8_0 +blk.10.ffn_up.weight q8_0 +blk.11.attn_k.weight q8_0 +blk.11.attn_output.weight q8_0 +blk.11.attn_q.weight q8_0 +blk.11.attn_v.weight q8_0 +blk.11.ffn_down.weight q8_0 +blk.11.ffn_gate.weight q8_0 +blk.11.ffn_up.weight q8_0 +blk.12.attn_k.weight q8_0 +blk.12.attn_output.weight q8_0 +blk.12.attn_q.weight q8_0 +blk.12.attn_v.weight q8_0 +blk.12.ffn_down.weight q8_0 +blk.12.ffn_gate.weight q8_0 +blk.12.ffn_up.weight q8_0 +blk.13.attn_k.weight q8_0 +blk.13.attn_output.weight q8_0 +blk.13.attn_q.weight q8_0 +blk.13.attn_v.weight q8_0 +blk.13.ffn_down.weight q8_0 +blk.13.ffn_gate.weight q8_0 +blk.13.ffn_up.weight q8_0 +blk.14.attn_k.weight q8_0 +blk.14.attn_output.weight q8_0 +blk.14.attn_q.weight q8_0 +blk.14.attn_v.weight q8_0 blk.14.ffn_down.weight q8_0 -blk.15.ffn_down.weight q8_0 -blk.15.ffn_gate.weight q8_0 -blk.15.ffn_up.weight q8_0 +blk.14.ffn_gate.weight q8_0 +blk.14.ffn_up.weight q8_0 blk.15.attn_k.weight q8_0 blk.15.attn_output.weight q8_0 blk.15.attn_q.weight q8_0 blk.15.attn_v.weight q8_0 -blk.16.ffn_down.weight q8_0 -blk.16.ffn_gate.weight q8_0 -blk.16.ffn_up.weight q8_0 +blk.15.ffn_down.weight q8_0 +blk.15.ffn_gate.weight q8_0 +blk.15.ffn_up.weight q8_0 blk.16.attn_k.weight q8_0 blk.16.attn_output.weight q8_0 blk.16.attn_q.weight q8_0 blk.16.attn_v.weight q8_0 -blk.17.ffn_down.weight q8_0 -blk.17.ffn_gate.weight q8_0 -blk.17.ffn_up.weight q8_0 +blk.16.ffn_down.weight q8_0 +blk.16.ffn_gate.weight q8_0 +blk.16.ffn_up.weight q8_0 blk.17.attn_k.weight q8_0 blk.17.attn_output.weight q8_0 blk.17.attn_q.weight q8_0 blk.17.attn_v.weight q8_0 -blk.18.ffn_down.weight q8_0 -blk.18.ffn_gate.weight q8_0 -blk.18.ffn_up.weight q8_0 +blk.17.ffn_down.weight q8_0 +blk.17.ffn_gate.weight q8_0 +blk.17.ffn_up.weight q8_0 blk.18.attn_k.weight q8_0 blk.18.attn_output.weight q8_0 blk.18.attn_q.weight q8_0 blk.18.attn_v.weight q8_0 -blk.19.ffn_down.weight q8_0 -blk.19.ffn_gate.weight q8_0 -blk.19.ffn_up.weight q8_0 +blk.18.ffn_down.weight q8_0 +blk.18.ffn_gate.weight q8_0 +blk.18.ffn_up.weight q8_0 blk.19.attn_k.weight q8_0 blk.19.attn_output.weight q8_0 blk.19.attn_q.weight q8_0 blk.19.attn_v.weight q8_0 -blk.20.ffn_down.weight q8_0 -blk.20.ffn_gate.weight q8_0 -blk.20.ffn_up.weight q8_0 +blk.19.ffn_down.weight q8_0 +blk.19.ffn_gate.weight q8_0 +blk.19.ffn_up.weight q8_0 blk.20.attn_k.weight q8_0 blk.20.attn_output.weight q8_0 blk.20.attn_q.weight q8_0 blk.20.attn_v.weight q8_0 -blk.21.ffn_down.weight q8_0 -blk.21.ffn_gate.weight q8_0 -blk.21.ffn_up.weight q8_0 +blk.20.ffn_down.weight q8_0 +blk.20.ffn_gate.weight q8_0 +blk.20.ffn_up.weight q8_0 blk.21.attn_k.weight q8_0 blk.21.attn_output.weight q8_0 blk.21.attn_q.weight q8_0 blk.21.attn_v.weight q8_0 -blk.22.ffn_down.weight q8_0 -blk.22.ffn_gate.weight q8_0 -blk.22.ffn_up.weight q8_0 +blk.21.ffn_down.weight q8_0 +blk.21.ffn_gate.weight q8_0 +blk.21.ffn_up.weight q8_0 blk.22.attn_k.weight q8_0 blk.22.attn_output.weight q8_0 blk.22.attn_q.weight q8_0 blk.22.attn_v.weight q8_0 -blk.23.ffn_down.weight q8_0 -blk.23.ffn_gate.weight q8_0 -blk.23.ffn_up.weight q8_0 +blk.22.ffn_down.weight q8_0 +blk.22.ffn_gate.weight q8_0 +blk.22.ffn_up.weight q8_0 blk.23.attn_k.weight q8_0 blk.23.attn_output.weight q8_0 blk.23.attn_q.weight q8_0 blk.23.attn_v.weight q8_0 -blk.24.ffn_down.weight q8_0 -blk.24.ffn_gate.weight q8_0 -blk.24.ffn_up.weight q8_0 +blk.23.ffn_down.weight q8_0 +blk.23.ffn_gate.weight q8_0 +blk.23.ffn_up.weight q8_0 blk.24.attn_k.weight q8_0 blk.24.attn_output.weight q8_0 blk.24.attn_q.weight q8_0 blk.24.attn_v.weight q8_0 -blk.25.ffn_down.weight q8_0 -blk.25.ffn_gate.weight q8_0 -blk.25.ffn_up.weight q8_0 +blk.24.ffn_down.weight q8_0 +blk.24.ffn_gate.weight q8_0 +blk.24.ffn_up.weight q8_0 blk.25.attn_k.weight q8_0 blk.25.attn_output.weight q8_0 blk.25.attn_q.weight q8_0 blk.25.attn_v.weight q8_0 -blk.26.ffn_down.weight q8_0 -blk.26.ffn_gate.weight q8_0 -blk.26.ffn_up.weight q8_0 +blk.25.ffn_down.weight q8_0 +blk.25.ffn_gate.weight q8_0 +blk.25.ffn_up.weight q8_0 blk.26.attn_k.weight q8_0 blk.26.attn_output.weight q8_0 blk.26.attn_q.weight q8_0 blk.26.attn_v.weight q8_0 -blk.27.ffn_down.weight q8_0 -blk.27.ffn_gate.weight q8_0 -blk.27.ffn_up.weight q8_0 +blk.26.ffn_down.weight q8_0 +blk.26.ffn_gate.weight q8_0 +blk.26.ffn_up.weight q8_0 blk.27.attn_k.weight q8_0 blk.27.attn_output.weight q8_0 blk.27.attn_q.weight q8_0 blk.27.attn_v.weight q8_0 -blk.28.ffn_down.weight q8_0 -blk.28.ffn_gate.weight q8_0 -blk.28.ffn_up.weight q8_0 +blk.27.ffn_down.weight q8_0 +blk.27.ffn_gate.weight q8_0 +blk.27.ffn_up.weight q8_0 blk.28.attn_k.weight q8_0 blk.28.attn_output.weight q8_0 blk.28.attn_q.weight q8_0 blk.28.attn_v.weight q8_0 -blk.29.ffn_down.weight q8_0 -blk.29.ffn_gate.weight q8_0 -blk.29.ffn_up.weight q8_0 +blk.28.ffn_down.weight q8_0 +blk.28.ffn_gate.weight q8_0 +blk.28.ffn_up.weight q8_0 blk.29.attn_k.weight q8_0 blk.29.attn_output.weight q8_0 blk.29.attn_q.weight q8_0 blk.29.attn_v.weight q8_0 -blk.30.ffn_down.weight q8_0 -blk.30.ffn_gate.weight q8_0 -blk.30.ffn_up.weight q8_0 +blk.29.ffn_down.weight q8_0 +blk.29.ffn_gate.weight q8_0 +blk.29.ffn_up.weight q8_0 blk.30.attn_k.weight q8_0 blk.30.attn_output.weight q8_0 blk.30.attn_q.weight q8_0 blk.30.attn_v.weight q8_0 -blk.31.ffn_down.weight q8_0 -blk.31.ffn_gate.weight q8_0 -blk.31.ffn_up.weight q8_0 +blk.30.ffn_down.weight q8_0 +blk.30.ffn_gate.weight q8_0 +blk.30.ffn_up.weight q8_0 blk.31.attn_k.weight q8_0 blk.31.attn_output.weight q8_0 blk.31.attn_q.weight q8_0 blk.31.attn_v.weight q8_0 -blk.32.ffn_down.weight q8_0 -blk.32.ffn_gate.weight q8_0 -blk.32.ffn_up.weight q8_0 +blk.31.ffn_down.weight q8_0 +blk.31.ffn_gate.weight q8_0 +blk.31.ffn_up.weight q8_0 blk.32.attn_k.weight q8_0 blk.32.attn_output.weight q8_0 blk.32.attn_q.weight q8_0 blk.32.attn_v.weight q8_0 -blk.33.ffn_down.weight q8_0 -blk.33.ffn_gate.weight q8_0 -blk.33.ffn_up.weight q8_0 +blk.32.ffn_down.weight q8_0 +blk.32.ffn_gate.weight q8_0 +blk.32.ffn_up.weight q8_0 blk.33.attn_k.weight q8_0 blk.33.attn_output.weight q8_0 blk.33.attn_q.weight q8_0 blk.33.attn_v.weight q8_0 +blk.33.ffn_down.weight q8_0 +blk.33.ffn_gate.weight q8_0 +blk.33.ffn_up.weight q8_0 diff --git a/tests/snapshots/glm-4.6v.schema b/tests/snapshots/glm-4.6v.schema index 560745c0fd..ff892b6568 100644 --- a/tests/snapshots/glm-4.6v.schema +++ b/tests/snapshots/glm-4.6v.schema @@ -20,192 +20,193 @@ output.weight q6_K output.weight q6_K [Q2_K] q2_K -blk.0.ffn_down.weight q4_0 +output.weight q6_K blk.0.attn_output.weight q3_K blk.0.attn_v.weight q4_K -blk.1.ffn_down_exps.weight q4_0 -blk.1.ffn_down_shexp.weight q4_0 +blk.0.ffn_down.weight q4_0 blk.1.attn_output.weight q3_K blk.1.attn_v.weight q4_K -blk.2.ffn_down_exps.weight q4_0 -blk.2.ffn_down_shexp.weight q4_0 +blk.1.ffn_down_exps.weight q4_0 +blk.1.ffn_down_shexp.weight q4_0 blk.2.attn_output.weight q3_K blk.2.attn_v.weight q4_K -blk.3.ffn_down_exps.weight q4_0 -blk.3.ffn_down_shexp.weight q4_0 +blk.2.ffn_down_exps.weight q4_0 +blk.2.ffn_down_shexp.weight q4_0 blk.3.attn_output.weight q3_K blk.3.attn_v.weight q4_K -blk.4.ffn_down_exps.weight q4_0 -blk.4.ffn_down_shexp.weight q4_0 +blk.3.ffn_down_exps.weight q4_0 +blk.3.ffn_down_shexp.weight q4_0 blk.4.attn_output.weight q3_K blk.4.attn_v.weight q4_K -blk.5.ffn_down_exps.weight q4_0 -blk.5.ffn_down_shexp.weight q4_0 +blk.4.ffn_down_exps.weight q4_0 +blk.4.ffn_down_shexp.weight q4_0 blk.5.attn_output.weight q3_K blk.5.attn_v.weight q4_K -blk.6.ffn_down_exps.weight q4_0 -blk.6.ffn_down_shexp.weight q4_0 +blk.5.ffn_down_exps.weight q4_0 +blk.5.ffn_down_shexp.weight q4_0 blk.6.attn_output.weight q3_K blk.6.attn_v.weight q4_K -blk.7.ffn_down_exps.weight q4_0 -blk.7.ffn_down_shexp.weight q4_0 +blk.6.ffn_down_exps.weight q4_0 +blk.6.ffn_down_shexp.weight q4_0 blk.7.attn_output.weight q3_K blk.7.attn_v.weight q4_K -blk.8.ffn_down_exps.weight q4_0 -blk.8.ffn_down_shexp.weight q4_0 +blk.7.ffn_down_exps.weight q4_0 +blk.7.ffn_down_shexp.weight q4_0 blk.8.attn_output.weight q3_K blk.8.attn_v.weight q4_K -blk.9.ffn_down_exps.weight q4_0 -blk.9.ffn_down_shexp.weight q4_0 +blk.8.ffn_down_exps.weight q4_0 +blk.8.ffn_down_shexp.weight q4_0 blk.9.attn_output.weight q3_K blk.9.attn_v.weight q4_K -blk.10.ffn_down_exps.weight q4_0 -blk.10.ffn_down_shexp.weight q4_0 +blk.9.ffn_down_exps.weight q4_0 +blk.9.ffn_down_shexp.weight q4_0 blk.10.attn_output.weight q3_K blk.10.attn_v.weight q4_K -blk.11.ffn_down_exps.weight q4_0 -blk.11.ffn_down_shexp.weight q4_0 +blk.10.ffn_down_exps.weight q4_0 +blk.10.ffn_down_shexp.weight q4_0 blk.11.attn_output.weight q3_K blk.11.attn_v.weight q4_K -blk.12.ffn_down_exps.weight q4_0 -blk.12.ffn_down_shexp.weight q4_0 +blk.11.ffn_down_exps.weight q4_0 +blk.11.ffn_down_shexp.weight q4_0 blk.12.attn_output.weight q3_K blk.12.attn_v.weight q4_K -blk.13.ffn_down_exps.weight q4_0 -blk.13.ffn_down_shexp.weight q4_0 +blk.12.ffn_down_exps.weight q4_0 +blk.12.ffn_down_shexp.weight q4_0 blk.13.attn_output.weight q3_K blk.13.attn_v.weight q4_K -blk.14.ffn_down_exps.weight q4_0 -blk.14.ffn_down_shexp.weight q4_0 +blk.13.ffn_down_exps.weight q4_0 +blk.13.ffn_down_shexp.weight q4_0 blk.14.attn_output.weight q3_K blk.14.attn_v.weight q4_K -blk.15.ffn_down_exps.weight q4_0 -blk.15.ffn_down_shexp.weight q4_0 +blk.14.ffn_down_exps.weight q4_0 +blk.14.ffn_down_shexp.weight q4_0 blk.15.attn_output.weight q3_K blk.15.attn_v.weight q4_K -blk.16.ffn_down_exps.weight q4_0 -blk.16.ffn_down_shexp.weight q4_0 +blk.15.ffn_down_exps.weight q4_0 +blk.15.ffn_down_shexp.weight q4_0 blk.16.attn_output.weight q3_K blk.16.attn_v.weight q4_K -blk.17.ffn_down_exps.weight q4_0 -blk.17.ffn_down_shexp.weight q4_0 +blk.16.ffn_down_exps.weight q4_0 +blk.16.ffn_down_shexp.weight q4_0 blk.17.attn_output.weight q3_K blk.17.attn_v.weight q4_K -blk.18.ffn_down_exps.weight q4_0 -blk.18.ffn_down_shexp.weight q4_0 +blk.17.ffn_down_exps.weight q4_0 +blk.17.ffn_down_shexp.weight q4_0 blk.18.attn_output.weight q3_K blk.18.attn_v.weight q4_K -blk.19.ffn_down_exps.weight q4_0 -blk.19.ffn_down_shexp.weight q4_0 +blk.18.ffn_down_exps.weight q4_0 +blk.18.ffn_down_shexp.weight q4_0 blk.19.attn_output.weight q3_K blk.19.attn_v.weight q4_K -blk.20.ffn_down_exps.weight q4_0 -blk.20.ffn_down_shexp.weight q4_0 +blk.19.ffn_down_exps.weight q4_0 +blk.19.ffn_down_shexp.weight q4_0 blk.20.attn_output.weight q3_K blk.20.attn_v.weight q4_K -blk.21.ffn_down_exps.weight q4_0 -blk.21.ffn_down_shexp.weight q4_0 +blk.20.ffn_down_exps.weight q4_0 +blk.20.ffn_down_shexp.weight q4_0 blk.21.attn_output.weight q3_K blk.21.attn_v.weight q4_K -blk.22.ffn_down_exps.weight q4_0 -blk.22.ffn_down_shexp.weight q4_0 +blk.21.ffn_down_exps.weight q4_0 +blk.21.ffn_down_shexp.weight q4_0 blk.22.attn_output.weight q3_K blk.22.attn_v.weight q4_K -blk.23.ffn_down_exps.weight q4_0 -blk.23.ffn_down_shexp.weight q4_0 +blk.22.ffn_down_exps.weight q4_0 +blk.22.ffn_down_shexp.weight q4_0 blk.23.attn_output.weight q3_K blk.23.attn_v.weight q4_K -blk.24.ffn_down_exps.weight q4_0 -blk.24.ffn_down_shexp.weight q4_0 +blk.23.ffn_down_exps.weight q4_0 +blk.23.ffn_down_shexp.weight q4_0 blk.24.attn_output.weight q3_K blk.24.attn_v.weight q4_K -blk.25.ffn_down_exps.weight q4_0 -blk.25.ffn_down_shexp.weight q4_0 +blk.24.ffn_down_exps.weight q4_0 +blk.24.ffn_down_shexp.weight q4_0 blk.25.attn_output.weight q3_K blk.25.attn_v.weight q4_K -blk.26.ffn_down_exps.weight q4_0 -blk.26.ffn_down_shexp.weight q4_0 +blk.25.ffn_down_exps.weight q4_0 +blk.25.ffn_down_shexp.weight q4_0 blk.26.attn_output.weight q3_K blk.26.attn_v.weight q4_K -blk.27.ffn_down_exps.weight q4_0 -blk.27.ffn_down_shexp.weight q4_0 +blk.26.ffn_down_exps.weight q4_0 +blk.26.ffn_down_shexp.weight q4_0 blk.27.attn_output.weight q3_K blk.27.attn_v.weight q4_K -blk.28.ffn_down_exps.weight q4_0 -blk.28.ffn_down_shexp.weight q4_0 +blk.27.ffn_down_exps.weight q4_0 +blk.27.ffn_down_shexp.weight q4_0 blk.28.attn_output.weight q3_K blk.28.attn_v.weight q4_K -blk.29.ffn_down_exps.weight q4_0 -blk.29.ffn_down_shexp.weight q4_0 +blk.28.ffn_down_exps.weight q4_0 +blk.28.ffn_down_shexp.weight q4_0 blk.29.attn_output.weight q3_K blk.29.attn_v.weight q4_K -blk.30.ffn_down_exps.weight q4_0 -blk.30.ffn_down_shexp.weight q4_0 +blk.29.ffn_down_exps.weight q4_0 +blk.29.ffn_down_shexp.weight q4_0 blk.30.attn_output.weight q3_K blk.30.attn_v.weight q4_K -blk.31.ffn_down_exps.weight q4_0 -blk.31.ffn_down_shexp.weight q4_0 +blk.30.ffn_down_exps.weight q4_0 +blk.30.ffn_down_shexp.weight q4_0 blk.31.attn_output.weight q3_K blk.31.attn_v.weight q4_K -blk.32.ffn_down_exps.weight q4_0 -blk.32.ffn_down_shexp.weight q4_0 +blk.31.ffn_down_exps.weight q4_0 +blk.31.ffn_down_shexp.weight q4_0 blk.32.attn_output.weight q3_K blk.32.attn_v.weight q4_K -blk.33.ffn_down_exps.weight q4_0 -blk.33.ffn_down_shexp.weight q4_0 +blk.32.ffn_down_exps.weight q4_0 +blk.32.ffn_down_shexp.weight q4_0 blk.33.attn_output.weight q3_K blk.33.attn_v.weight q4_K -blk.34.ffn_down_exps.weight q4_0 +blk.33.ffn_down_exps.weight q4_0 +blk.33.ffn_down_shexp.weight q4_0 blk.34.attn_output.weight q3_K -blk.34.ffn_down_shexp.weight q4_0 blk.34.attn_v.weight q4_K -blk.35.ffn_down_exps.weight q4_0 -blk.35.ffn_down_shexp.weight q4_0 +blk.34.ffn_down_exps.weight q4_0 +blk.34.ffn_down_shexp.weight q4_0 blk.35.attn_output.weight q3_K blk.35.attn_v.weight q4_K -blk.36.ffn_down_exps.weight q4_0 -blk.36.ffn_down_shexp.weight q4_0 +blk.35.ffn_down_exps.weight q4_0 +blk.35.ffn_down_shexp.weight q4_0 blk.36.attn_output.weight q3_K blk.36.attn_v.weight q4_K -blk.37.ffn_down_exps.weight q4_0 -blk.37.ffn_down_shexp.weight q4_0 +blk.36.ffn_down_exps.weight q4_0 +blk.36.ffn_down_shexp.weight q4_0 blk.37.attn_output.weight q3_K blk.37.attn_v.weight q4_K -blk.38.ffn_down_exps.weight q4_0 -blk.38.ffn_down_shexp.weight q4_0 +blk.37.ffn_down_exps.weight q4_0 +blk.37.ffn_down_shexp.weight q4_0 blk.38.attn_output.weight q3_K blk.38.attn_v.weight q4_K -blk.39.ffn_down_exps.weight q4_0 -blk.39.ffn_down_shexp.weight q4_0 +blk.38.ffn_down_exps.weight q4_0 +blk.38.ffn_down_shexp.weight q4_0 blk.39.attn_output.weight q3_K blk.39.attn_v.weight q4_K -blk.40.ffn_down_exps.weight q4_0 -blk.40.ffn_down_shexp.weight q4_0 +blk.39.ffn_down_exps.weight q4_0 +blk.39.ffn_down_shexp.weight q4_0 blk.40.attn_output.weight q3_K blk.40.attn_v.weight q4_K -blk.41.ffn_down_exps.weight q4_0 -blk.41.ffn_down_shexp.weight q4_0 +blk.40.ffn_down_exps.weight q4_0 +blk.40.ffn_down_shexp.weight q4_0 blk.41.attn_output.weight q3_K blk.41.attn_v.weight q4_K +blk.41.ffn_down_exps.weight q4_0 +blk.41.ffn_down_shexp.weight q4_0 +blk.42.attn_output.weight q3_K +blk.42.attn_v.weight q4_K blk.42.ffn_down_exps.weight q4_0 blk.42.ffn_down_shexp.weight q4_0 -blk.42.attn_v.weight q4_K -blk.42.attn_output.weight q3_K -blk.43.ffn_down_exps.weight q4_0 -blk.43.ffn_down_shexp.weight q4_0 blk.43.attn_output.weight q3_K blk.43.attn_v.weight q4_K -blk.44.ffn_down_exps.weight q4_0 -blk.44.ffn_down_shexp.weight q4_0 +blk.43.ffn_down_exps.weight q4_0 +blk.43.ffn_down_shexp.weight q4_0 blk.44.attn_output.weight q3_K blk.44.attn_v.weight q4_K -output.weight q6_K -blk.45.ffn_down_exps.weight q4_0 -blk.45.ffn_down_shexp.weight q4_0 +blk.44.ffn_down_exps.weight q4_0 +blk.44.ffn_down_shexp.weight q4_0 blk.45.attn_output.weight q3_K blk.45.attn_v.weight q4_K +blk.45.ffn_down_exps.weight q4_0 +blk.45.ffn_down_shexp.weight q4_0 [Q3_K_S] q3_K +output.weight q6_K blk.0.ffn_down.weight q4_0 blk.1.ffn_down_exps.weight q4_0 blk.1.ffn_down_shexp.weight q4_0 @@ -295,394 +296,394 @@ blk.43.ffn_down_exps.weight q4_0 blk.43.ffn_down_shexp.weight q4_0 blk.44.ffn_down_exps.weight q4_0 blk.44.ffn_down_shexp.weight q4_0 -output.weight q6_K blk.45.ffn_down_exps.weight q4_0 blk.45.ffn_down_shexp.weight q4_0 [Q3_K_M] q3_K -blk.0.ffn_down.weight q5_1 +output.weight q6_K blk.0.attn_output.weight q4_K blk.0.attn_v.weight q5_K -blk.1.ffn_down_exps.weight q5_1 -blk.1.ffn_down_shexp.weight q5_1 +blk.0.ffn_down.weight q5_1 blk.1.attn_output.weight q4_K blk.1.attn_v.weight q5_K -blk.2.ffn_down_exps.weight q5_0 -blk.2.ffn_down_shexp.weight q5_0 +blk.1.ffn_down_exps.weight q5_1 +blk.1.ffn_down_shexp.weight q5_1 blk.2.attn_output.weight q4_K blk.2.attn_v.weight q4_K -blk.3.ffn_down_exps.weight q5_0 -blk.3.ffn_down_shexp.weight q5_0 +blk.2.ffn_down_exps.weight q5_0 +blk.2.ffn_down_shexp.weight q5_0 blk.3.attn_output.weight q4_K blk.3.attn_v.weight q4_K -blk.4.ffn_down_exps.weight q5_0 -blk.4.ffn_down_shexp.weight q5_0 +blk.3.ffn_down_exps.weight q5_0 +blk.3.ffn_down_shexp.weight q5_0 blk.4.attn_output.weight q4_K blk.4.attn_v.weight q4_K -blk.5.ffn_down_exps.weight q5_0 -blk.5.ffn_down_shexp.weight q5_0 +blk.4.ffn_down_exps.weight q5_0 +blk.4.ffn_down_shexp.weight q5_0 blk.5.attn_output.weight q4_K blk.5.attn_v.weight q4_K -blk.6.ffn_down_exps.weight q5_0 -blk.6.ffn_down_shexp.weight q5_0 +blk.5.ffn_down_exps.weight q5_0 +blk.5.ffn_down_shexp.weight q5_0 blk.6.attn_output.weight q4_K blk.6.attn_v.weight q4_K -blk.7.ffn_down_exps.weight q5_0 -blk.7.ffn_down_shexp.weight q5_0 +blk.6.ffn_down_exps.weight q5_0 +blk.6.ffn_down_shexp.weight q5_0 blk.7.attn_output.weight q4_K blk.7.attn_v.weight q4_K -blk.8.ffn_down_exps.weight q5_0 -blk.8.ffn_down_shexp.weight q5_0 +blk.7.ffn_down_exps.weight q5_0 +blk.7.ffn_down_shexp.weight q5_0 blk.8.attn_output.weight q4_K blk.8.attn_v.weight q4_K -blk.9.ffn_down_exps.weight q5_0 -blk.9.ffn_down_shexp.weight q5_0 +blk.8.ffn_down_exps.weight q5_0 +blk.8.ffn_down_shexp.weight q5_0 blk.9.attn_output.weight q4_K blk.9.attn_v.weight q4_K -blk.10.ffn_down_exps.weight q5_0 -blk.10.ffn_down_shexp.weight q5_0 +blk.9.ffn_down_exps.weight q5_0 +blk.9.ffn_down_shexp.weight q5_0 blk.10.attn_output.weight q4_K blk.10.attn_v.weight q4_K -blk.11.ffn_down_exps.weight q5_0 -blk.11.ffn_down_shexp.weight q5_0 +blk.10.ffn_down_exps.weight q5_0 +blk.10.ffn_down_shexp.weight q5_0 blk.11.attn_output.weight q4_K blk.11.attn_v.weight q4_K -blk.12.ffn_down_exps.weight q5_0 -blk.12.ffn_down_shexp.weight q5_0 +blk.11.ffn_down_exps.weight q5_0 +blk.11.ffn_down_shexp.weight q5_0 blk.12.attn_output.weight q4_K blk.12.attn_v.weight q4_K -blk.13.ffn_down_exps.weight q5_0 -blk.13.ffn_down_shexp.weight q5_0 +blk.12.ffn_down_exps.weight q5_0 +blk.12.ffn_down_shexp.weight q5_0 blk.13.attn_output.weight q4_K blk.13.attn_v.weight q4_K -blk.14.ffn_down_exps.weight q5_0 -blk.14.ffn_down_shexp.weight q5_0 +blk.13.ffn_down_exps.weight q5_0 +blk.13.ffn_down_shexp.weight q5_0 blk.14.attn_output.weight q4_K blk.14.attn_v.weight q4_K -blk.15.ffn_down_exps.weight q5_0 -blk.15.ffn_down_shexp.weight q5_0 +blk.14.ffn_down_exps.weight q5_0 +blk.14.ffn_down_shexp.weight q5_0 blk.15.attn_output.weight q4_K blk.15.attn_v.weight q4_K -blk.16.ffn_down_exps.weight q5_0 -blk.16.ffn_down_shexp.weight q5_0 +blk.15.ffn_down_exps.weight q5_0 +blk.15.ffn_down_shexp.weight q5_0 blk.16.attn_output.weight q4_K blk.16.attn_v.weight q4_K -blk.17.ffn_down_exps.weight q5_0 -blk.17.ffn_down_shexp.weight q5_0 +blk.16.ffn_down_exps.weight q5_0 +blk.16.ffn_down_shexp.weight q5_0 blk.17.attn_output.weight q4_K blk.17.attn_v.weight q4_K -blk.18.ffn_down_exps.weight q5_0 -blk.18.ffn_down_shexp.weight q5_0 +blk.17.ffn_down_exps.weight q5_0 +blk.17.ffn_down_shexp.weight q5_0 blk.18.attn_output.weight q4_K blk.18.attn_v.weight q4_K -blk.19.ffn_down_exps.weight q5_0 -blk.19.ffn_down_shexp.weight q5_0 +blk.18.ffn_down_exps.weight q5_0 +blk.18.ffn_down_shexp.weight q5_0 blk.19.attn_output.weight q4_K blk.19.attn_v.weight q4_K -blk.20.ffn_down_exps.weight q5_0 -blk.20.ffn_down_shexp.weight q5_0 +blk.19.ffn_down_exps.weight q5_0 +blk.19.ffn_down_shexp.weight q5_0 blk.20.attn_output.weight q4_K blk.20.attn_v.weight q4_K -blk.21.ffn_down_exps.weight q5_0 -blk.21.ffn_down_shexp.weight q5_0 +blk.20.ffn_down_exps.weight q5_0 +blk.20.ffn_down_shexp.weight q5_0 blk.21.attn_output.weight q4_K blk.21.attn_v.weight q4_K -blk.22.ffn_down_exps.weight q5_0 -blk.22.ffn_down_shexp.weight q5_0 +blk.21.ffn_down_exps.weight q5_0 +blk.21.ffn_down_shexp.weight q5_0 blk.22.attn_output.weight q4_K blk.22.attn_v.weight q4_K -blk.23.ffn_down_exps.weight q5_0 -blk.23.ffn_down_shexp.weight q5_0 +blk.22.ffn_down_exps.weight q5_0 +blk.22.ffn_down_shexp.weight q5_0 blk.23.attn_output.weight q4_K blk.23.attn_v.weight q4_K -blk.24.ffn_down_exps.weight q5_0 -blk.24.ffn_down_shexp.weight q5_0 +blk.23.ffn_down_exps.weight q5_0 +blk.23.ffn_down_shexp.weight q5_0 blk.24.attn_output.weight q4_K blk.24.attn_v.weight q4_K -blk.25.ffn_down_exps.weight q5_0 -blk.25.ffn_down_shexp.weight q5_0 +blk.24.ffn_down_exps.weight q5_0 +blk.24.ffn_down_shexp.weight q5_0 blk.25.attn_output.weight q4_K blk.25.attn_v.weight q4_K -blk.26.ffn_down_exps.weight q5_0 -blk.26.ffn_down_shexp.weight q5_0 +blk.25.ffn_down_exps.weight q5_0 +blk.25.ffn_down_shexp.weight q5_0 blk.26.attn_output.weight q4_K blk.26.attn_v.weight q4_K -blk.27.ffn_down_exps.weight q5_0 -blk.27.ffn_down_shexp.weight q5_0 +blk.26.ffn_down_exps.weight q5_0 +blk.26.ffn_down_shexp.weight q5_0 blk.27.attn_output.weight q4_K blk.27.attn_v.weight q4_K -blk.28.ffn_down_exps.weight q5_0 -blk.28.ffn_down_shexp.weight q5_0 +blk.27.ffn_down_exps.weight q5_0 +blk.27.ffn_down_shexp.weight q5_0 blk.28.attn_output.weight q4_K blk.28.attn_v.weight q4_K -blk.29.ffn_down_exps.weight q5_0 -blk.29.ffn_down_shexp.weight q5_0 +blk.28.ffn_down_exps.weight q5_0 +blk.28.ffn_down_shexp.weight q5_0 blk.29.attn_output.weight q4_K blk.29.attn_v.weight q4_K -blk.30.ffn_down_exps.weight q5_0 -blk.30.ffn_down_shexp.weight q5_0 +blk.29.ffn_down_exps.weight q5_0 +blk.29.ffn_down_shexp.weight q5_0 blk.30.attn_output.weight q4_K blk.30.attn_v.weight q4_K -blk.31.ffn_down_exps.weight q5_0 -blk.31.ffn_down_shexp.weight q5_0 +blk.30.ffn_down_exps.weight q5_0 +blk.30.ffn_down_shexp.weight q5_0 blk.31.attn_output.weight q4_K blk.31.attn_v.weight q4_K -blk.32.ffn_down_exps.weight q5_0 -blk.32.ffn_down_shexp.weight q5_0 +blk.31.ffn_down_exps.weight q5_0 +blk.31.ffn_down_shexp.weight q5_0 blk.32.attn_output.weight q4_K blk.32.attn_v.weight q4_K -blk.33.ffn_down_exps.weight q5_0 -blk.33.ffn_down_shexp.weight q5_0 +blk.32.ffn_down_exps.weight q5_0 +blk.32.ffn_down_shexp.weight q5_0 blk.33.attn_output.weight q4_K blk.33.attn_v.weight q4_K -blk.34.ffn_down_exps.weight q5_0 +blk.33.ffn_down_exps.weight q5_0 +blk.33.ffn_down_shexp.weight q5_0 blk.34.attn_output.weight q4_K -blk.34.ffn_down_shexp.weight q5_0 blk.34.attn_v.weight q4_K -blk.35.ffn_down_exps.weight q5_0 -blk.35.ffn_down_shexp.weight q5_0 +blk.34.ffn_down_exps.weight q5_0 +blk.34.ffn_down_shexp.weight q5_0 blk.35.attn_output.weight q4_K blk.35.attn_v.weight q4_K -blk.36.ffn_down_exps.weight q5_0 -blk.36.ffn_down_shexp.weight q5_0 +blk.35.ffn_down_exps.weight q5_0 +blk.35.ffn_down_shexp.weight q5_0 blk.36.attn_output.weight q4_K blk.36.attn_v.weight q4_K -blk.37.ffn_down_exps.weight q5_0 -blk.37.ffn_down_shexp.weight q5_0 +blk.36.ffn_down_exps.weight q5_0 +blk.36.ffn_down_shexp.weight q5_0 blk.37.attn_output.weight q4_K blk.37.attn_v.weight q4_K -blk.38.ffn_down_exps.weight q5_0 -blk.38.ffn_down_shexp.weight q5_0 +blk.37.ffn_down_exps.weight q5_0 +blk.37.ffn_down_shexp.weight q5_0 blk.38.attn_output.weight q4_K blk.38.attn_v.weight q4_K -blk.39.ffn_down_exps.weight q5_0 -blk.39.ffn_down_shexp.weight q5_0 +blk.38.ffn_down_exps.weight q5_0 +blk.38.ffn_down_shexp.weight q5_0 blk.39.attn_output.weight q4_K blk.39.attn_v.weight q4_K -blk.40.ffn_down_exps.weight q5_0 -blk.40.ffn_down_shexp.weight q5_0 +blk.39.ffn_down_exps.weight q5_0 +blk.39.ffn_down_shexp.weight q5_0 blk.40.attn_output.weight q4_K blk.40.attn_v.weight q4_K -blk.41.ffn_down_exps.weight q5_0 -blk.41.ffn_down_shexp.weight q5_0 +blk.40.ffn_down_exps.weight q5_0 +blk.40.ffn_down_shexp.weight q5_0 blk.41.attn_output.weight q4_K blk.41.attn_v.weight q4_K +blk.41.ffn_down_exps.weight q5_0 +blk.41.ffn_down_shexp.weight q5_0 +blk.42.attn_output.weight q4_K +blk.42.attn_v.weight q4_K blk.42.ffn_down_exps.weight q5_0 blk.42.ffn_down_shexp.weight q5_0 -blk.42.attn_v.weight q4_K -blk.42.attn_output.weight q4_K -blk.43.ffn_down_exps.weight q5_0 -blk.43.ffn_down_shexp.weight q5_0 blk.43.attn_output.weight q4_K blk.43.attn_v.weight q4_K -blk.44.ffn_down_exps.weight q5_0 -blk.44.ffn_down_shexp.weight q5_0 +blk.43.ffn_down_exps.weight q5_0 +blk.43.ffn_down_shexp.weight q5_0 blk.44.attn_output.weight q4_K blk.44.attn_v.weight q4_K -output.weight q6_K -blk.45.ffn_down_exps.weight q5_0 -blk.45.ffn_down_shexp.weight q5_0 +blk.44.ffn_down_exps.weight q5_0 +blk.44.ffn_down_shexp.weight q5_0 blk.45.attn_output.weight q4_K blk.45.attn_v.weight q4_K +blk.45.ffn_down_exps.weight q5_0 +blk.45.ffn_down_shexp.weight q5_0 [Q3_K_L] q3_K -blk.0.ffn_down.weight q5_1 +output.weight q6_K blk.0.attn_output.weight q5_K blk.0.attn_v.weight q5_K -blk.1.ffn_down_exps.weight q5_1 -blk.1.ffn_down_shexp.weight q5_1 +blk.0.ffn_down.weight q5_1 blk.1.attn_output.weight q5_K blk.1.attn_v.weight q5_K -blk.2.ffn_down_exps.weight q5_1 -blk.2.ffn_down_shexp.weight q5_1 -blk.2.attn_output.weight q5_K -blk.2.attn_v.weight q5_K -blk.3.ffn_down_exps.weight q5_1 -blk.3.ffn_down_shexp.weight q5_1 -blk.3.attn_output.weight q5_K -blk.3.attn_v.weight q5_K -blk.4.ffn_down_exps.weight q5_1 -blk.4.ffn_down_shexp.weight q5_1 -blk.4.attn_output.weight q5_K -blk.4.attn_v.weight q5_K -blk.5.ffn_down_exps.weight q5_1 -blk.5.ffn_down_shexp.weight q5_1 -blk.5.attn_output.weight q5_K -blk.5.attn_v.weight q5_K -blk.6.ffn_down_exps.weight q5_1 -blk.6.ffn_down_shexp.weight q5_1 -blk.6.attn_output.weight q5_K -blk.6.attn_v.weight q5_K -blk.7.ffn_down_exps.weight q5_1 -blk.7.ffn_down_shexp.weight q5_1 -blk.7.attn_output.weight q5_K -blk.7.attn_v.weight q5_K -blk.8.ffn_down_exps.weight q5_1 -blk.8.ffn_down_shexp.weight q5_1 -blk.8.attn_output.weight q5_K -blk.8.attn_v.weight q5_K -blk.9.ffn_down_exps.weight q5_1 -blk.9.ffn_down_shexp.weight q5_1 -blk.9.attn_output.weight q5_K -blk.9.attn_v.weight q5_K -blk.10.ffn_down_exps.weight q5_1 -blk.10.ffn_down_shexp.weight q5_1 -blk.10.attn_output.weight q5_K -blk.10.attn_v.weight q5_K -blk.11.ffn_down_exps.weight q5_1 -blk.11.ffn_down_shexp.weight q5_1 -blk.11.attn_output.weight q5_K -blk.11.attn_v.weight q5_K -blk.12.ffn_down_exps.weight q5_1 -blk.12.ffn_down_shexp.weight q5_1 -blk.12.attn_output.weight q5_K -blk.12.attn_v.weight q5_K -blk.13.ffn_down_exps.weight q5_1 -blk.13.ffn_down_shexp.weight q5_1 -blk.13.attn_output.weight q5_K -blk.13.attn_v.weight q5_K -blk.14.ffn_down_exps.weight q5_1 -blk.14.ffn_down_shexp.weight q5_1 -blk.14.attn_output.weight q5_K -blk.14.attn_v.weight q5_K -blk.15.ffn_down_exps.weight q5_1 -blk.15.ffn_down_shexp.weight q5_1 -blk.15.attn_output.weight q5_K -blk.15.attn_v.weight q5_K -blk.16.ffn_down_exps.weight q5_1 -blk.16.ffn_down_shexp.weight q5_1 -blk.16.attn_output.weight q5_K -blk.16.attn_v.weight q5_K -blk.17.ffn_down_exps.weight q5_1 -blk.17.ffn_down_shexp.weight q5_1 -blk.17.attn_output.weight q5_K -blk.17.attn_v.weight q5_K -blk.18.ffn_down_exps.weight q5_1 -blk.18.ffn_down_shexp.weight q5_1 -blk.18.attn_output.weight q5_K -blk.18.attn_v.weight q5_K -blk.19.ffn_down_exps.weight q5_1 -blk.19.ffn_down_shexp.weight q5_1 -blk.19.attn_output.weight q5_K -blk.19.attn_v.weight q5_K -blk.20.ffn_down_exps.weight q5_1 -blk.20.ffn_down_shexp.weight q5_1 -blk.20.attn_output.weight q5_K -blk.20.attn_v.weight q5_K -blk.21.ffn_down_exps.weight q5_1 -blk.21.ffn_down_shexp.weight q5_1 -blk.21.attn_output.weight q5_K -blk.21.attn_v.weight q5_K -blk.22.ffn_down_exps.weight q5_1 -blk.22.ffn_down_shexp.weight q5_1 -blk.22.attn_output.weight q5_K -blk.22.attn_v.weight q5_K -blk.23.ffn_down_exps.weight q5_1 -blk.23.ffn_down_shexp.weight q5_1 -blk.23.attn_output.weight q5_K -blk.23.attn_v.weight q5_K -blk.24.ffn_down_exps.weight q5_1 -blk.24.ffn_down_shexp.weight q5_1 -blk.24.attn_output.weight q5_K -blk.24.attn_v.weight q5_K -blk.25.ffn_down_exps.weight q5_1 -blk.25.ffn_down_shexp.weight q5_1 -blk.25.attn_output.weight q5_K -blk.25.attn_v.weight q5_K -blk.26.ffn_down_exps.weight q5_1 -blk.26.ffn_down_shexp.weight q5_1 -blk.26.attn_output.weight q5_K -blk.26.attn_v.weight q5_K -blk.27.ffn_down_exps.weight q5_1 -blk.27.ffn_down_shexp.weight q5_1 -blk.27.attn_output.weight q5_K -blk.27.attn_v.weight q5_K -blk.28.ffn_down_exps.weight q5_1 -blk.28.ffn_down_shexp.weight q5_1 -blk.28.attn_output.weight q5_K -blk.28.attn_v.weight q5_K -blk.29.ffn_down_exps.weight q5_1 -blk.29.ffn_down_shexp.weight q5_1 -blk.29.attn_output.weight q5_K -blk.29.attn_v.weight q5_K -blk.30.ffn_down_exps.weight q5_1 -blk.30.ffn_down_shexp.weight q5_1 -blk.30.attn_output.weight q5_K -blk.30.attn_v.weight q5_K -blk.31.ffn_down_exps.weight q5_1 -blk.31.ffn_down_shexp.weight q5_1 -blk.31.attn_output.weight q5_K -blk.31.attn_v.weight q5_K -blk.32.ffn_down_exps.weight q5_1 -blk.32.ffn_down_shexp.weight q5_1 -blk.32.attn_output.weight q5_K -blk.32.attn_v.weight q5_K -blk.33.ffn_down_exps.weight q5_1 -blk.33.ffn_down_shexp.weight q5_1 -blk.33.attn_output.weight q5_K -blk.33.attn_v.weight q5_K -blk.34.ffn_down_exps.weight q5_1 -blk.34.attn_output.weight q5_K -blk.34.ffn_down_shexp.weight q5_1 -blk.34.attn_v.weight q5_K -blk.35.ffn_down_exps.weight q5_1 -blk.35.ffn_down_shexp.weight q5_1 -blk.35.attn_output.weight q5_K -blk.35.attn_v.weight q5_K -blk.36.ffn_down_exps.weight q5_1 -blk.36.ffn_down_shexp.weight q5_1 -blk.36.attn_output.weight q5_K -blk.36.attn_v.weight q5_K -blk.37.ffn_down_exps.weight q5_1 -blk.37.ffn_down_shexp.weight q5_1 -blk.37.attn_output.weight q5_K -blk.37.attn_v.weight q5_K -blk.38.ffn_down_exps.weight q5_1 -blk.38.ffn_down_shexp.weight q5_1 -blk.38.attn_output.weight q5_K -blk.38.attn_v.weight q5_K -blk.39.ffn_down_exps.weight q5_1 -blk.39.ffn_down_shexp.weight q5_1 -blk.39.attn_output.weight q5_K -blk.39.attn_v.weight q5_K -blk.40.ffn_down_exps.weight q5_1 -blk.40.ffn_down_shexp.weight q5_1 -blk.40.attn_output.weight q5_K -blk.40.attn_v.weight q5_K -blk.41.ffn_down_exps.weight q5_1 -blk.41.ffn_down_shexp.weight q5_1 -blk.41.attn_output.weight q5_K -blk.41.attn_v.weight q5_K -blk.42.ffn_down_exps.weight q5_1 -blk.42.ffn_down_shexp.weight q5_1 -blk.42.attn_v.weight q5_K -blk.42.attn_output.weight q5_K -blk.43.ffn_down_exps.weight q5_1 -blk.43.ffn_down_shexp.weight q5_1 -blk.43.attn_output.weight q5_K -blk.43.attn_v.weight q5_K -blk.44.ffn_down_exps.weight q5_1 -blk.44.ffn_down_shexp.weight q5_1 -blk.44.attn_output.weight q5_K -blk.44.attn_v.weight q5_K -output.weight q6_K -blk.45.ffn_down_exps.weight q5_1 -blk.45.ffn_down_shexp.weight q5_1 -blk.45.attn_output.weight q5_K -blk.45.attn_v.weight q5_K - -[Q4_K_S] q4_K -blk.0.ffn_down.weight q5_1 -blk.0.attn_v.weight q5_K blk.1.ffn_down_exps.weight q5_1 blk.1.ffn_down_shexp.weight q5_1 -blk.1.attn_v.weight q5_K +blk.2.attn_output.weight q5_K +blk.2.attn_v.weight q5_K blk.2.ffn_down_exps.weight q5_1 blk.2.ffn_down_shexp.weight q5_1 -blk.2.attn_v.weight q5_K +blk.3.attn_output.weight q5_K +blk.3.attn_v.weight q5_K blk.3.ffn_down_exps.weight q5_1 blk.3.ffn_down_shexp.weight q5_1 +blk.4.attn_output.weight q5_K +blk.4.attn_v.weight q5_K +blk.4.ffn_down_exps.weight q5_1 +blk.4.ffn_down_shexp.weight q5_1 +blk.5.attn_output.weight q5_K +blk.5.attn_v.weight q5_K +blk.5.ffn_down_exps.weight q5_1 +blk.5.ffn_down_shexp.weight q5_1 +blk.6.attn_output.weight q5_K +blk.6.attn_v.weight q5_K +blk.6.ffn_down_exps.weight q5_1 +blk.6.ffn_down_shexp.weight q5_1 +blk.7.attn_output.weight q5_K +blk.7.attn_v.weight q5_K +blk.7.ffn_down_exps.weight q5_1 +blk.7.ffn_down_shexp.weight q5_1 +blk.8.attn_output.weight q5_K +blk.8.attn_v.weight q5_K +blk.8.ffn_down_exps.weight q5_1 +blk.8.ffn_down_shexp.weight q5_1 +blk.9.attn_output.weight q5_K +blk.9.attn_v.weight q5_K +blk.9.ffn_down_exps.weight q5_1 +blk.9.ffn_down_shexp.weight q5_1 +blk.10.attn_output.weight q5_K +blk.10.attn_v.weight q5_K +blk.10.ffn_down_exps.weight q5_1 +blk.10.ffn_down_shexp.weight q5_1 +blk.11.attn_output.weight q5_K +blk.11.attn_v.weight q5_K +blk.11.ffn_down_exps.weight q5_1 +blk.11.ffn_down_shexp.weight q5_1 +blk.12.attn_output.weight q5_K +blk.12.attn_v.weight q5_K +blk.12.ffn_down_exps.weight q5_1 +blk.12.ffn_down_shexp.weight q5_1 +blk.13.attn_output.weight q5_K +blk.13.attn_v.weight q5_K +blk.13.ffn_down_exps.weight q5_1 +blk.13.ffn_down_shexp.weight q5_1 +blk.14.attn_output.weight q5_K +blk.14.attn_v.weight q5_K +blk.14.ffn_down_exps.weight q5_1 +blk.14.ffn_down_shexp.weight q5_1 +blk.15.attn_output.weight q5_K +blk.15.attn_v.weight q5_K +blk.15.ffn_down_exps.weight q5_1 +blk.15.ffn_down_shexp.weight q5_1 +blk.16.attn_output.weight q5_K +blk.16.attn_v.weight q5_K +blk.16.ffn_down_exps.weight q5_1 +blk.16.ffn_down_shexp.weight q5_1 +blk.17.attn_output.weight q5_K +blk.17.attn_v.weight q5_K +blk.17.ffn_down_exps.weight q5_1 +blk.17.ffn_down_shexp.weight q5_1 +blk.18.attn_output.weight q5_K +blk.18.attn_v.weight q5_K +blk.18.ffn_down_exps.weight q5_1 +blk.18.ffn_down_shexp.weight q5_1 +blk.19.attn_output.weight q5_K +blk.19.attn_v.weight q5_K +blk.19.ffn_down_exps.weight q5_1 +blk.19.ffn_down_shexp.weight q5_1 +blk.20.attn_output.weight q5_K +blk.20.attn_v.weight q5_K +blk.20.ffn_down_exps.weight q5_1 +blk.20.ffn_down_shexp.weight q5_1 +blk.21.attn_output.weight q5_K +blk.21.attn_v.weight q5_K +blk.21.ffn_down_exps.weight q5_1 +blk.21.ffn_down_shexp.weight q5_1 +blk.22.attn_output.weight q5_K +blk.22.attn_v.weight q5_K +blk.22.ffn_down_exps.weight q5_1 +blk.22.ffn_down_shexp.weight q5_1 +blk.23.attn_output.weight q5_K +blk.23.attn_v.weight q5_K +blk.23.ffn_down_exps.weight q5_1 +blk.23.ffn_down_shexp.weight q5_1 +blk.24.attn_output.weight q5_K +blk.24.attn_v.weight q5_K +blk.24.ffn_down_exps.weight q5_1 +blk.24.ffn_down_shexp.weight q5_1 +blk.25.attn_output.weight q5_K +blk.25.attn_v.weight q5_K +blk.25.ffn_down_exps.weight q5_1 +blk.25.ffn_down_shexp.weight q5_1 +blk.26.attn_output.weight q5_K +blk.26.attn_v.weight q5_K +blk.26.ffn_down_exps.weight q5_1 +blk.26.ffn_down_shexp.weight q5_1 +blk.27.attn_output.weight q5_K +blk.27.attn_v.weight q5_K +blk.27.ffn_down_exps.weight q5_1 +blk.27.ffn_down_shexp.weight q5_1 +blk.28.attn_output.weight q5_K +blk.28.attn_v.weight q5_K +blk.28.ffn_down_exps.weight q5_1 +blk.28.ffn_down_shexp.weight q5_1 +blk.29.attn_output.weight q5_K +blk.29.attn_v.weight q5_K +blk.29.ffn_down_exps.weight q5_1 +blk.29.ffn_down_shexp.weight q5_1 +blk.30.attn_output.weight q5_K +blk.30.attn_v.weight q5_K +blk.30.ffn_down_exps.weight q5_1 +blk.30.ffn_down_shexp.weight q5_1 +blk.31.attn_output.weight q5_K +blk.31.attn_v.weight q5_K +blk.31.ffn_down_exps.weight q5_1 +blk.31.ffn_down_shexp.weight q5_1 +blk.32.attn_output.weight q5_K +blk.32.attn_v.weight q5_K +blk.32.ffn_down_exps.weight q5_1 +blk.32.ffn_down_shexp.weight q5_1 +blk.33.attn_output.weight q5_K +blk.33.attn_v.weight q5_K +blk.33.ffn_down_exps.weight q5_1 +blk.33.ffn_down_shexp.weight q5_1 +blk.34.attn_output.weight q5_K +blk.34.attn_v.weight q5_K +blk.34.ffn_down_exps.weight q5_1 +blk.34.ffn_down_shexp.weight q5_1 +blk.35.attn_output.weight q5_K +blk.35.attn_v.weight q5_K +blk.35.ffn_down_exps.weight q5_1 +blk.35.ffn_down_shexp.weight q5_1 +blk.36.attn_output.weight q5_K +blk.36.attn_v.weight q5_K +blk.36.ffn_down_exps.weight q5_1 +blk.36.ffn_down_shexp.weight q5_1 +blk.37.attn_output.weight q5_K +blk.37.attn_v.weight q5_K +blk.37.ffn_down_exps.weight q5_1 +blk.37.ffn_down_shexp.weight q5_1 +blk.38.attn_output.weight q5_K +blk.38.attn_v.weight q5_K +blk.38.ffn_down_exps.weight q5_1 +blk.38.ffn_down_shexp.weight q5_1 +blk.39.attn_output.weight q5_K +blk.39.attn_v.weight q5_K +blk.39.ffn_down_exps.weight q5_1 +blk.39.ffn_down_shexp.weight q5_1 +blk.40.attn_output.weight q5_K +blk.40.attn_v.weight q5_K +blk.40.ffn_down_exps.weight q5_1 +blk.40.ffn_down_shexp.weight q5_1 +blk.41.attn_output.weight q5_K +blk.41.attn_v.weight q5_K +blk.41.ffn_down_exps.weight q5_1 +blk.41.ffn_down_shexp.weight q5_1 +blk.42.attn_output.weight q5_K +blk.42.attn_v.weight q5_K +blk.42.ffn_down_exps.weight q5_1 +blk.42.ffn_down_shexp.weight q5_1 +blk.43.attn_output.weight q5_K +blk.43.attn_v.weight q5_K +blk.43.ffn_down_exps.weight q5_1 +blk.43.ffn_down_shexp.weight q5_1 +blk.44.attn_output.weight q5_K +blk.44.attn_v.weight q5_K +blk.44.ffn_down_exps.weight q5_1 +blk.44.ffn_down_shexp.weight q5_1 +blk.45.attn_output.weight q5_K +blk.45.attn_v.weight q5_K +blk.45.ffn_down_exps.weight q5_1 +blk.45.ffn_down_shexp.weight q5_1 + +[Q4_K_S] q4_K +output.weight q6_K +blk.0.attn_v.weight q5_K +blk.0.ffn_down.weight q5_1 +blk.1.attn_v.weight q5_K +blk.1.ffn_down_exps.weight q5_1 +blk.1.ffn_down_shexp.weight q5_1 +blk.2.attn_v.weight q5_K +blk.2.ffn_down_exps.weight q5_1 +blk.2.ffn_down_shexp.weight q5_1 blk.3.attn_v.weight q5_K +blk.3.ffn_down_exps.weight q5_1 +blk.3.ffn_down_shexp.weight q5_1 blk.4.ffn_down_exps.weight q5_1 blk.4.ffn_down_shexp.weight q5_1 blk.5.ffn_down_exps.weight q5_0 @@ -765,127 +766,127 @@ blk.43.ffn_down_exps.weight q5_0 blk.43.ffn_down_shexp.weight q5_0 blk.44.ffn_down_exps.weight q5_0 blk.44.ffn_down_shexp.weight q5_0 -output.weight q6_K blk.45.ffn_down_exps.weight q5_0 blk.45.ffn_down_shexp.weight q5_0 [Q4_K_M] q4_K -blk.0.ffn_down.weight q8_0 +output.weight q6_K blk.0.attn_v.weight q6_K +blk.0.ffn_down.weight q8_0 +blk.1.attn_v.weight q6_K blk.1.ffn_down_exps.weight q8_0 blk.1.ffn_down_shexp.weight q8_0 -blk.1.attn_v.weight q6_K +blk.2.attn_v.weight q6_K blk.2.ffn_down_exps.weight q8_0 blk.2.ffn_down_shexp.weight q8_0 -blk.2.attn_v.weight q6_K +blk.3.attn_v.weight q6_K blk.3.ffn_down_exps.weight q8_0 blk.3.ffn_down_shexp.weight q8_0 -blk.3.attn_v.weight q6_K +blk.4.attn_v.weight q6_K blk.4.ffn_down_exps.weight q8_0 blk.4.ffn_down_shexp.weight q8_0 -blk.4.attn_v.weight q6_K blk.5.ffn_down_exps.weight q5_0 blk.5.ffn_down_shexp.weight q5_0 blk.6.ffn_down_exps.weight q5_0 blk.6.ffn_down_shexp.weight q5_0 +blk.7.attn_v.weight q6_K blk.7.ffn_down_exps.weight q8_0 blk.7.ffn_down_shexp.weight q8_0 -blk.7.attn_v.weight q6_K blk.8.ffn_down_exps.weight q5_0 blk.8.ffn_down_shexp.weight q5_0 blk.9.ffn_down_exps.weight q5_0 blk.9.ffn_down_shexp.weight q5_0 +blk.10.attn_v.weight q6_K blk.10.ffn_down_exps.weight q8_0 blk.10.ffn_down_shexp.weight q8_0 -blk.10.attn_v.weight q6_K blk.11.ffn_down_exps.weight q5_0 blk.11.ffn_down_shexp.weight q5_0 blk.12.ffn_down_exps.weight q5_0 blk.12.ffn_down_shexp.weight q5_0 +blk.13.attn_v.weight q6_K blk.13.ffn_down_exps.weight q8_0 blk.13.ffn_down_shexp.weight q8_0 -blk.13.attn_v.weight q6_K blk.14.ffn_down_exps.weight q5_0 blk.14.ffn_down_shexp.weight q5_0 blk.15.ffn_down_exps.weight q5_0 blk.15.ffn_down_shexp.weight q5_0 +blk.16.attn_v.weight q6_K blk.16.ffn_down_exps.weight q8_0 blk.16.ffn_down_shexp.weight q8_0 -blk.16.attn_v.weight q6_K blk.17.ffn_down_exps.weight q5_0 blk.17.ffn_down_shexp.weight q5_0 blk.18.ffn_down_exps.weight q5_0 blk.18.ffn_down_shexp.weight q5_0 +blk.19.attn_v.weight q6_K blk.19.ffn_down_exps.weight q8_0 blk.19.ffn_down_shexp.weight q8_0 -blk.19.attn_v.weight q6_K blk.20.ffn_down_exps.weight q5_0 blk.20.ffn_down_shexp.weight q5_0 blk.21.ffn_down_exps.weight q5_0 blk.21.ffn_down_shexp.weight q5_0 +blk.22.attn_v.weight q6_K blk.22.ffn_down_exps.weight q8_0 blk.22.ffn_down_shexp.weight q8_0 -blk.22.attn_v.weight q6_K blk.23.ffn_down_exps.weight q5_0 blk.23.ffn_down_shexp.weight q5_0 blk.24.ffn_down_exps.weight q5_0 blk.24.ffn_down_shexp.weight q5_0 +blk.25.attn_v.weight q6_K blk.25.ffn_down_exps.weight q8_0 blk.25.ffn_down_shexp.weight q8_0 -blk.25.attn_v.weight q6_K blk.26.ffn_down_exps.weight q5_0 blk.26.ffn_down_shexp.weight q5_0 blk.27.ffn_down_exps.weight q5_0 blk.27.ffn_down_shexp.weight q5_0 +blk.28.attn_v.weight q6_K blk.28.ffn_down_exps.weight q8_0 blk.28.ffn_down_shexp.weight q8_0 -blk.28.attn_v.weight q6_K blk.29.ffn_down_exps.weight q5_0 blk.29.ffn_down_shexp.weight q5_0 blk.30.ffn_down_exps.weight q5_0 blk.30.ffn_down_shexp.weight q5_0 +blk.31.attn_v.weight q6_K blk.31.ffn_down_exps.weight q8_0 blk.31.ffn_down_shexp.weight q8_0 -blk.31.attn_v.weight q6_K blk.32.ffn_down_exps.weight q5_0 blk.32.ffn_down_shexp.weight q5_0 blk.33.ffn_down_exps.weight q5_0 blk.33.ffn_down_shexp.weight q5_0 +blk.34.attn_v.weight q6_K blk.34.ffn_down_exps.weight q8_0 blk.34.ffn_down_shexp.weight q8_0 -blk.34.attn_v.weight q6_K blk.35.ffn_down_exps.weight q5_0 blk.35.ffn_down_shexp.weight q5_0 blk.36.ffn_down_exps.weight q5_0 blk.36.ffn_down_shexp.weight q5_0 +blk.37.attn_v.weight q6_K blk.37.ffn_down_exps.weight q8_0 blk.37.ffn_down_shexp.weight q8_0 -blk.37.attn_v.weight q6_K blk.38.ffn_down_exps.weight q5_0 blk.38.ffn_down_shexp.weight q5_0 blk.39.ffn_down_exps.weight q5_0 blk.39.ffn_down_shexp.weight q5_0 +blk.40.attn_v.weight q6_K blk.40.ffn_down_exps.weight q8_0 blk.40.ffn_down_shexp.weight q8_0 -blk.40.attn_v.weight q6_K +blk.41.attn_v.weight q6_K blk.41.ffn_down_exps.weight q8_0 blk.41.ffn_down_shexp.weight q8_0 -blk.41.attn_v.weight q6_K +blk.42.attn_v.weight q6_K blk.42.ffn_down_exps.weight q8_0 blk.42.ffn_down_shexp.weight q8_0 -blk.42.attn_v.weight q6_K +blk.43.attn_v.weight q6_K blk.43.ffn_down_exps.weight q8_0 blk.43.ffn_down_shexp.weight q8_0 -blk.43.attn_v.weight q6_K +blk.44.attn_v.weight q6_K blk.44.ffn_down_exps.weight q8_0 blk.44.ffn_down_shexp.weight q8_0 -blk.44.attn_v.weight q6_K -output.weight q6_K +blk.45.attn_v.weight q6_K blk.45.ffn_down_exps.weight q8_0 blk.45.ffn_down_shexp.weight q8_0 -blk.45.attn_v.weight q6_K [Q5_K_S] q5_K +output.weight q6_K blk.0.ffn_down.weight q5_1 blk.1.ffn_down_exps.weight q5_1 blk.1.ffn_down_shexp.weight q5_1 @@ -975,125 +976,124 @@ blk.43.ffn_down_exps.weight q5_1 blk.43.ffn_down_shexp.weight q5_1 blk.44.ffn_down_exps.weight q5_1 blk.44.ffn_down_shexp.weight q5_1 -output.weight q6_K blk.45.ffn_down_exps.weight q5_1 blk.45.ffn_down_shexp.weight q5_1 [Q5_K_M] q5_K -blk.0.ffn_down.weight q8_0 +output.weight q6_K blk.0.attn_v.weight q6_K +blk.0.ffn_down.weight q8_0 +blk.1.attn_v.weight q6_K blk.1.ffn_down_exps.weight q8_0 blk.1.ffn_down_shexp.weight q8_0 -blk.1.attn_v.weight q6_K +blk.2.attn_v.weight q6_K blk.2.ffn_down_exps.weight q8_0 blk.2.ffn_down_shexp.weight q8_0 -blk.2.attn_v.weight q6_K +blk.3.attn_v.weight q6_K blk.3.ffn_down_exps.weight q8_0 blk.3.ffn_down_shexp.weight q8_0 -blk.3.attn_v.weight q6_K +blk.4.attn_v.weight q6_K blk.4.ffn_down_exps.weight q8_0 blk.4.ffn_down_shexp.weight q8_0 -blk.4.attn_v.weight q6_K blk.5.ffn_down_exps.weight q5_1 blk.5.ffn_down_shexp.weight q5_1 blk.6.ffn_down_exps.weight q5_1 blk.6.ffn_down_shexp.weight q5_1 +blk.7.attn_v.weight q6_K blk.7.ffn_down_exps.weight q8_0 blk.7.ffn_down_shexp.weight q8_0 -blk.7.attn_v.weight q6_K blk.8.ffn_down_exps.weight q5_1 blk.8.ffn_down_shexp.weight q5_1 blk.9.ffn_down_exps.weight q5_1 blk.9.ffn_down_shexp.weight q5_1 +blk.10.attn_v.weight q6_K blk.10.ffn_down_exps.weight q8_0 blk.10.ffn_down_shexp.weight q8_0 -blk.10.attn_v.weight q6_K blk.11.ffn_down_exps.weight q5_1 blk.11.ffn_down_shexp.weight q5_1 blk.12.ffn_down_exps.weight q5_1 blk.12.ffn_down_shexp.weight q5_1 +blk.13.attn_v.weight q6_K blk.13.ffn_down_exps.weight q8_0 blk.13.ffn_down_shexp.weight q8_0 -blk.13.attn_v.weight q6_K blk.14.ffn_down_exps.weight q5_1 blk.14.ffn_down_shexp.weight q5_1 blk.15.ffn_down_exps.weight q5_1 blk.15.ffn_down_shexp.weight q5_1 +blk.16.attn_v.weight q6_K blk.16.ffn_down_exps.weight q8_0 blk.16.ffn_down_shexp.weight q8_0 -blk.16.attn_v.weight q6_K blk.17.ffn_down_exps.weight q5_1 blk.17.ffn_down_shexp.weight q5_1 blk.18.ffn_down_exps.weight q5_1 blk.18.ffn_down_shexp.weight q5_1 +blk.19.attn_v.weight q6_K blk.19.ffn_down_exps.weight q8_0 blk.19.ffn_down_shexp.weight q8_0 -blk.19.attn_v.weight q6_K blk.20.ffn_down_exps.weight q5_1 blk.20.ffn_down_shexp.weight q5_1 blk.21.ffn_down_exps.weight q5_1 blk.21.ffn_down_shexp.weight q5_1 +blk.22.attn_v.weight q6_K blk.22.ffn_down_exps.weight q8_0 blk.22.ffn_down_shexp.weight q8_0 -blk.22.attn_v.weight q6_K blk.23.ffn_down_exps.weight q5_1 blk.23.ffn_down_shexp.weight q5_1 blk.24.ffn_down_exps.weight q5_1 blk.24.ffn_down_shexp.weight q5_1 +blk.25.attn_v.weight q6_K blk.25.ffn_down_exps.weight q8_0 blk.25.ffn_down_shexp.weight q8_0 -blk.25.attn_v.weight q6_K blk.26.ffn_down_exps.weight q5_1 blk.26.ffn_down_shexp.weight q5_1 blk.27.ffn_down_exps.weight q5_1 blk.27.ffn_down_shexp.weight q5_1 +blk.28.attn_v.weight q6_K blk.28.ffn_down_exps.weight q8_0 blk.28.ffn_down_shexp.weight q8_0 -blk.28.attn_v.weight q6_K blk.29.ffn_down_exps.weight q5_1 blk.29.ffn_down_shexp.weight q5_1 blk.30.ffn_down_exps.weight q5_1 blk.30.ffn_down_shexp.weight q5_1 +blk.31.attn_v.weight q6_K blk.31.ffn_down_exps.weight q8_0 blk.31.ffn_down_shexp.weight q8_0 -blk.31.attn_v.weight q6_K blk.32.ffn_down_exps.weight q5_1 blk.32.ffn_down_shexp.weight q5_1 blk.33.ffn_down_exps.weight q5_1 blk.33.ffn_down_shexp.weight q5_1 +blk.34.attn_v.weight q6_K blk.34.ffn_down_exps.weight q8_0 blk.34.ffn_down_shexp.weight q8_0 -blk.34.attn_v.weight q6_K blk.35.ffn_down_exps.weight q5_1 blk.35.ffn_down_shexp.weight q5_1 blk.36.ffn_down_exps.weight q5_1 blk.36.ffn_down_shexp.weight q5_1 +blk.37.attn_v.weight q6_K blk.37.ffn_down_exps.weight q8_0 blk.37.ffn_down_shexp.weight q8_0 -blk.37.attn_v.weight q6_K blk.38.ffn_down_exps.weight q5_1 blk.38.ffn_down_shexp.weight q5_1 blk.39.ffn_down_exps.weight q5_1 blk.39.ffn_down_shexp.weight q5_1 +blk.40.attn_v.weight q6_K blk.40.ffn_down_exps.weight q8_0 blk.40.ffn_down_shexp.weight q8_0 -blk.40.attn_v.weight q6_K +blk.41.attn_v.weight q6_K blk.41.ffn_down_exps.weight q8_0 blk.41.ffn_down_shexp.weight q8_0 -blk.41.attn_v.weight q6_K +blk.42.attn_v.weight q6_K blk.42.ffn_down_exps.weight q8_0 blk.42.ffn_down_shexp.weight q8_0 -blk.42.attn_v.weight q6_K +blk.43.attn_v.weight q6_K blk.43.ffn_down_exps.weight q8_0 blk.43.ffn_down_shexp.weight q8_0 -blk.43.attn_v.weight q6_K +blk.44.attn_v.weight q6_K blk.44.ffn_down_exps.weight q8_0 blk.44.ffn_down_shexp.weight q8_0 -blk.44.attn_v.weight q6_K -output.weight q6_K +blk.45.attn_v.weight q6_K blk.45.ffn_down_exps.weight q8_0 blk.45.ffn_down_shexp.weight q8_0 -blk.45.attn_v.weight q6_K [Q6_K] q6_K blk.0.ffn_down.weight q8_0 @@ -1189,1280 +1189,1281 @@ blk.45.ffn_down_exps.weight q8_0 blk.45.ffn_down_shexp.weight q8_0 [IQ2_XXS] iq2_xxs +output.weight q5_K token_embd.weight q2_K -blk.0.ffn_down.weight q4_0 blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q4_0 +blk.1.attn_v.weight q4_K blk.1.ffn_down_exps.weight q4_0 blk.1.ffn_down_shexp.weight q4_0 -blk.1.attn_v.weight q4_K +blk.2.attn_v.weight q4_K blk.2.ffn_down_exps.weight q4_0 blk.2.ffn_down_shexp.weight q4_0 -blk.2.attn_v.weight q4_K +blk.3.attn_v.weight q4_K blk.3.ffn_down_exps.weight iq4_nl blk.3.ffn_down_shexp.weight iq4_nl -blk.3.attn_v.weight q4_K +blk.4.attn_v.weight q4_K blk.4.ffn_down_exps.weight iq4_nl blk.4.ffn_down_shexp.weight iq4_nl -blk.4.attn_v.weight q4_K +blk.5.attn_v.weight q4_K blk.5.ffn_down_exps.weight iq4_nl blk.5.ffn_down_shexp.weight iq4_nl -blk.5.attn_v.weight q4_K +blk.6.attn_v.weight q4_K blk.6.ffn_down_exps.weight iq4_nl blk.6.ffn_down_shexp.weight iq4_nl -blk.6.attn_v.weight q4_K +blk.7.attn_v.weight q4_K blk.7.ffn_down_exps.weight iq4_nl blk.7.ffn_down_shexp.weight iq4_nl -blk.7.attn_v.weight q4_K +blk.8.attn_v.weight q4_K blk.8.ffn_down_exps.weight iq4_nl blk.8.ffn_down_shexp.weight iq4_nl -blk.8.attn_v.weight q4_K +blk.9.attn_v.weight q4_K blk.9.ffn_down_exps.weight iq4_nl blk.9.ffn_down_shexp.weight iq4_nl -blk.9.attn_v.weight q4_K +blk.10.attn_v.weight q4_K blk.10.ffn_down_exps.weight iq4_nl blk.10.ffn_down_shexp.weight iq4_nl -blk.10.attn_v.weight q4_K +blk.11.attn_v.weight q4_K blk.11.ffn_down_exps.weight iq4_nl blk.11.ffn_down_shexp.weight iq4_nl -blk.11.attn_v.weight q4_K +blk.12.attn_v.weight q4_K blk.12.ffn_down_exps.weight iq4_nl blk.12.ffn_down_shexp.weight iq4_nl -blk.12.attn_v.weight q4_K +blk.13.attn_v.weight q4_K blk.13.ffn_down_exps.weight iq4_nl blk.13.ffn_down_shexp.weight iq4_nl -blk.13.attn_v.weight q4_K +blk.14.attn_v.weight q4_K blk.14.ffn_down_exps.weight iq4_nl blk.14.ffn_down_shexp.weight iq4_nl -blk.14.attn_v.weight q4_K +blk.15.attn_v.weight q4_K blk.15.ffn_down_exps.weight iq4_nl blk.15.ffn_down_shexp.weight iq4_nl -blk.15.attn_v.weight q4_K +blk.16.attn_v.weight q4_K blk.16.ffn_down_exps.weight iq4_nl blk.16.ffn_down_shexp.weight iq4_nl -blk.16.attn_v.weight q4_K +blk.17.attn_v.weight q4_K blk.17.ffn_down_exps.weight iq4_nl blk.17.ffn_down_shexp.weight iq4_nl -blk.17.attn_v.weight q4_K +blk.18.attn_v.weight q4_K blk.18.ffn_down_exps.weight iq4_nl blk.18.ffn_down_shexp.weight iq4_nl -blk.18.attn_v.weight q4_K +blk.19.attn_v.weight q4_K blk.19.ffn_down_exps.weight iq4_nl blk.19.ffn_down_shexp.weight iq4_nl -blk.19.attn_v.weight q4_K +blk.20.attn_v.weight q4_K blk.20.ffn_down_exps.weight iq4_nl blk.20.ffn_down_shexp.weight iq4_nl -blk.20.attn_v.weight q4_K +blk.21.attn_v.weight q4_K blk.21.ffn_down_exps.weight iq4_nl blk.21.ffn_down_shexp.weight iq4_nl -blk.21.attn_v.weight q4_K +blk.22.attn_v.weight q4_K blk.22.ffn_down_exps.weight iq4_nl blk.22.ffn_down_shexp.weight iq4_nl -blk.22.attn_v.weight q4_K +blk.23.attn_v.weight q4_K blk.23.ffn_down_exps.weight iq4_nl blk.23.ffn_down_shexp.weight iq4_nl -blk.23.attn_v.weight q4_K +blk.24.attn_v.weight q4_K blk.24.ffn_down_exps.weight iq4_nl blk.24.ffn_down_shexp.weight iq4_nl -blk.24.attn_v.weight q4_K +blk.25.attn_v.weight q4_K blk.25.ffn_down_exps.weight iq4_nl blk.25.ffn_down_shexp.weight iq4_nl -blk.25.attn_v.weight q4_K +blk.26.attn_v.weight q4_K blk.26.ffn_down_exps.weight iq4_nl blk.26.ffn_down_shexp.weight iq4_nl -blk.26.attn_v.weight q4_K +blk.27.attn_v.weight q4_K blk.27.ffn_down_exps.weight iq4_nl blk.27.ffn_down_shexp.weight iq4_nl -blk.27.attn_v.weight q4_K +blk.28.attn_v.weight q4_K blk.28.ffn_down_exps.weight iq4_nl blk.28.ffn_down_shexp.weight iq4_nl -blk.28.attn_v.weight q4_K +blk.29.attn_v.weight q4_K blk.29.ffn_down_exps.weight iq4_nl blk.29.ffn_down_shexp.weight iq4_nl -blk.29.attn_v.weight q4_K +blk.30.attn_v.weight q4_K blk.30.ffn_down_exps.weight iq4_nl blk.30.ffn_down_shexp.weight iq4_nl -blk.30.attn_v.weight q4_K +blk.31.attn_v.weight q4_K blk.31.ffn_down_exps.weight iq4_nl blk.31.ffn_down_shexp.weight iq4_nl -blk.31.attn_v.weight q4_K +blk.32.attn_v.weight q4_K blk.32.ffn_down_exps.weight iq4_nl blk.32.ffn_down_shexp.weight iq4_nl -blk.32.attn_v.weight q4_K +blk.33.attn_v.weight q4_K blk.33.ffn_down_exps.weight iq4_nl blk.33.ffn_down_shexp.weight iq4_nl -blk.33.attn_v.weight q4_K +blk.34.attn_v.weight q4_K blk.34.ffn_down_exps.weight iq4_nl blk.34.ffn_down_shexp.weight iq4_nl -blk.34.attn_v.weight q4_K +blk.35.attn_v.weight q4_K blk.35.ffn_down_exps.weight iq4_nl blk.35.ffn_down_shexp.weight iq4_nl -blk.35.attn_v.weight q4_K +blk.36.attn_v.weight q4_K blk.36.ffn_down_exps.weight iq4_nl blk.36.ffn_down_shexp.weight iq4_nl -blk.36.attn_v.weight q4_K +blk.37.attn_v.weight q4_K blk.37.ffn_down_exps.weight iq4_nl blk.37.ffn_down_shexp.weight iq4_nl -blk.37.attn_v.weight q4_K +blk.38.attn_v.weight q4_K blk.38.ffn_down_exps.weight iq4_nl blk.38.ffn_down_shexp.weight iq4_nl -blk.38.attn_v.weight q4_K +blk.39.attn_v.weight q4_K blk.39.ffn_down_exps.weight iq4_nl blk.39.ffn_down_shexp.weight iq4_nl -blk.39.attn_v.weight q4_K +blk.40.attn_v.weight q4_K blk.40.ffn_down_exps.weight iq4_nl blk.40.ffn_down_shexp.weight iq4_nl -blk.40.attn_v.weight q4_K +blk.41.attn_v.weight q4_K blk.41.ffn_down_exps.weight iq4_nl blk.41.ffn_down_shexp.weight iq4_nl -blk.41.attn_v.weight q4_K +blk.42.attn_v.weight q4_K blk.42.ffn_down_exps.weight iq4_nl blk.42.ffn_down_shexp.weight iq4_nl -blk.42.attn_v.weight q4_K +blk.43.attn_v.weight q4_K blk.43.ffn_down_exps.weight iq4_nl blk.43.ffn_down_shexp.weight iq4_nl -blk.43.attn_v.weight q4_K +blk.44.attn_v.weight q4_K blk.44.ffn_down_exps.weight iq4_nl blk.44.ffn_down_shexp.weight iq4_nl -blk.44.attn_v.weight q4_K -output.weight q5_K +blk.45.attn_v.weight q4_K blk.45.ffn_down_exps.weight iq4_nl blk.45.ffn_down_shexp.weight iq4_nl -blk.45.attn_v.weight q4_K [IQ2_XS] iq2_xs +output.weight q5_K token_embd.weight q2_K -blk.0.ffn_down.weight q4_0 blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q4_0 +blk.1.attn_v.weight q4_K blk.1.ffn_down_exps.weight q4_0 blk.1.ffn_down_shexp.weight q4_0 -blk.1.attn_v.weight q4_K +blk.2.attn_v.weight q4_K blk.2.ffn_down_exps.weight q4_0 blk.2.ffn_down_shexp.weight q4_0 -blk.2.attn_v.weight q4_K +blk.3.attn_v.weight q4_K blk.3.ffn_down_exps.weight iq4_nl blk.3.ffn_down_shexp.weight iq4_nl -blk.3.attn_v.weight q4_K +blk.4.attn_v.weight q4_K blk.4.ffn_down_exps.weight iq4_nl blk.4.ffn_down_shexp.weight iq4_nl -blk.4.attn_v.weight q4_K +blk.5.attn_v.weight q4_K blk.5.ffn_down_exps.weight iq4_nl blk.5.ffn_down_shexp.weight iq4_nl -blk.5.attn_v.weight q4_K +blk.6.attn_v.weight q4_K blk.6.ffn_down_exps.weight iq4_nl blk.6.ffn_down_shexp.weight iq4_nl -blk.6.attn_v.weight q4_K +blk.7.attn_v.weight q4_K blk.7.ffn_down_exps.weight iq4_nl blk.7.ffn_down_shexp.weight iq4_nl -blk.7.attn_v.weight q4_K +blk.8.attn_v.weight q4_K blk.8.ffn_down_exps.weight iq4_nl blk.8.ffn_down_shexp.weight iq4_nl -blk.8.attn_v.weight q4_K +blk.9.attn_v.weight q4_K blk.9.ffn_down_exps.weight iq4_nl blk.9.ffn_down_shexp.weight iq4_nl -blk.9.attn_v.weight q4_K +blk.10.attn_v.weight q4_K blk.10.ffn_down_exps.weight iq4_nl blk.10.ffn_down_shexp.weight iq4_nl -blk.10.attn_v.weight q4_K +blk.11.attn_v.weight q4_K blk.11.ffn_down_exps.weight iq4_nl blk.11.ffn_down_shexp.weight iq4_nl -blk.11.attn_v.weight q4_K +blk.12.attn_v.weight q4_K blk.12.ffn_down_exps.weight iq4_nl blk.12.ffn_down_shexp.weight iq4_nl -blk.12.attn_v.weight q4_K +blk.13.attn_v.weight q4_K blk.13.ffn_down_exps.weight iq4_nl blk.13.ffn_down_shexp.weight iq4_nl -blk.13.attn_v.weight q4_K +blk.14.attn_v.weight q4_K blk.14.ffn_down_exps.weight iq4_nl blk.14.ffn_down_shexp.weight iq4_nl -blk.14.attn_v.weight q4_K +blk.15.attn_v.weight q4_K blk.15.ffn_down_exps.weight iq4_nl blk.15.ffn_down_shexp.weight iq4_nl -blk.15.attn_v.weight q4_K +blk.16.attn_v.weight q4_K blk.16.ffn_down_exps.weight iq4_nl blk.16.ffn_down_shexp.weight iq4_nl -blk.16.attn_v.weight q4_K +blk.17.attn_v.weight q4_K blk.17.ffn_down_exps.weight iq4_nl blk.17.ffn_down_shexp.weight iq4_nl -blk.17.attn_v.weight q4_K +blk.18.attn_v.weight q4_K blk.18.ffn_down_exps.weight iq4_nl blk.18.ffn_down_shexp.weight iq4_nl -blk.18.attn_v.weight q4_K +blk.19.attn_v.weight q4_K blk.19.ffn_down_exps.weight iq4_nl blk.19.ffn_down_shexp.weight iq4_nl -blk.19.attn_v.weight q4_K +blk.20.attn_v.weight q4_K blk.20.ffn_down_exps.weight iq4_nl blk.20.ffn_down_shexp.weight iq4_nl -blk.20.attn_v.weight q4_K +blk.21.attn_v.weight q4_K blk.21.ffn_down_exps.weight iq4_nl blk.21.ffn_down_shexp.weight iq4_nl -blk.21.attn_v.weight q4_K +blk.22.attn_v.weight q4_K blk.22.ffn_down_exps.weight iq4_nl blk.22.ffn_down_shexp.weight iq4_nl -blk.22.attn_v.weight q4_K +blk.23.attn_v.weight q4_K blk.23.ffn_down_exps.weight iq4_nl blk.23.ffn_down_shexp.weight iq4_nl -blk.23.attn_v.weight q4_K +blk.24.attn_v.weight q4_K blk.24.ffn_down_exps.weight iq4_nl blk.24.ffn_down_shexp.weight iq4_nl -blk.24.attn_v.weight q4_K +blk.25.attn_v.weight q4_K blk.25.ffn_down_exps.weight iq4_nl blk.25.ffn_down_shexp.weight iq4_nl -blk.25.attn_v.weight q4_K +blk.26.attn_v.weight q4_K blk.26.ffn_down_exps.weight iq4_nl blk.26.ffn_down_shexp.weight iq4_nl -blk.26.attn_v.weight q4_K +blk.27.attn_v.weight q4_K blk.27.ffn_down_exps.weight iq4_nl blk.27.ffn_down_shexp.weight iq4_nl -blk.27.attn_v.weight q4_K +blk.28.attn_v.weight q4_K blk.28.ffn_down_exps.weight iq4_nl blk.28.ffn_down_shexp.weight iq4_nl -blk.28.attn_v.weight q4_K +blk.29.attn_v.weight q4_K blk.29.ffn_down_exps.weight iq4_nl blk.29.ffn_down_shexp.weight iq4_nl -blk.29.attn_v.weight q4_K +blk.30.attn_v.weight q4_K blk.30.ffn_down_exps.weight iq4_nl blk.30.ffn_down_shexp.weight iq4_nl -blk.30.attn_v.weight q4_K +blk.31.attn_v.weight q4_K blk.31.ffn_down_exps.weight iq4_nl blk.31.ffn_down_shexp.weight iq4_nl -blk.31.attn_v.weight q4_K +blk.32.attn_v.weight q4_K blk.32.ffn_down_exps.weight iq4_nl blk.32.ffn_down_shexp.weight iq4_nl -blk.32.attn_v.weight q4_K +blk.33.attn_v.weight q4_K blk.33.ffn_down_exps.weight iq4_nl blk.33.ffn_down_shexp.weight iq4_nl -blk.33.attn_v.weight q4_K +blk.34.attn_v.weight q4_K blk.34.ffn_down_exps.weight iq4_nl blk.34.ffn_down_shexp.weight iq4_nl -blk.34.attn_v.weight q4_K +blk.35.attn_v.weight q4_K blk.35.ffn_down_exps.weight iq4_nl blk.35.ffn_down_shexp.weight iq4_nl -blk.35.attn_v.weight q4_K +blk.36.attn_v.weight q4_K blk.36.ffn_down_exps.weight iq4_nl blk.36.ffn_down_shexp.weight iq4_nl -blk.36.attn_v.weight q4_K +blk.37.attn_v.weight q4_K blk.37.ffn_down_exps.weight iq4_nl blk.37.ffn_down_shexp.weight iq4_nl -blk.37.attn_v.weight q4_K +blk.38.attn_v.weight q4_K blk.38.ffn_down_exps.weight iq4_nl blk.38.ffn_down_shexp.weight iq4_nl -blk.38.attn_v.weight q4_K +blk.39.attn_v.weight q4_K blk.39.ffn_down_exps.weight iq4_nl blk.39.ffn_down_shexp.weight iq4_nl -blk.39.attn_v.weight q4_K +blk.40.attn_v.weight q4_K blk.40.ffn_down_exps.weight iq4_nl blk.40.ffn_down_shexp.weight iq4_nl -blk.40.attn_v.weight q4_K +blk.41.attn_v.weight q4_K blk.41.ffn_down_exps.weight iq4_nl blk.41.ffn_down_shexp.weight iq4_nl -blk.41.attn_v.weight q4_K +blk.42.attn_v.weight q4_K blk.42.ffn_down_exps.weight iq4_nl blk.42.ffn_down_shexp.weight iq4_nl -blk.42.attn_v.weight q4_K +blk.43.attn_v.weight q4_K blk.43.ffn_down_exps.weight iq4_nl blk.43.ffn_down_shexp.weight iq4_nl -blk.43.attn_v.weight q4_K +blk.44.attn_v.weight q4_K blk.44.ffn_down_exps.weight iq4_nl blk.44.ffn_down_shexp.weight iq4_nl -blk.44.attn_v.weight q4_K -output.weight q5_K +blk.45.attn_v.weight q4_K blk.45.ffn_down_exps.weight iq4_nl blk.45.ffn_down_shexp.weight iq4_nl -blk.45.attn_v.weight q4_K [Q2_K_S] q2_K -blk.0.ffn_down.weight q5_0 +output.weight q6_K blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q5_0 +blk.1.attn_v.weight q4_K blk.1.ffn_down_exps.weight q5_0 blk.1.ffn_down_shexp.weight q5_0 -blk.1.attn_v.weight q4_K +blk.2.attn_v.weight q4_K blk.2.ffn_down_exps.weight q5_0 blk.2.ffn_down_shexp.weight q5_0 -blk.2.attn_v.weight q4_K +blk.3.attn_v.weight q4_K blk.3.ffn_down_exps.weight q5_0 blk.3.ffn_down_shexp.weight q5_0 -blk.3.attn_v.weight q4_K +blk.4.attn_v.weight q4_K blk.4.ffn_down_exps.weight q5_0 blk.4.ffn_down_shexp.weight q5_0 -blk.4.attn_v.weight q4_K +blk.5.attn_v.weight q4_K blk.5.ffn_down_exps.weight q4_0 blk.5.ffn_down_shexp.weight q4_0 -blk.5.attn_v.weight q4_K +blk.6.attn_v.weight q4_K blk.6.ffn_down_exps.weight q4_0 blk.6.ffn_down_shexp.weight q4_0 -blk.6.attn_v.weight q4_K +blk.7.attn_v.weight q4_K blk.7.ffn_down_exps.weight q4_0 blk.7.ffn_down_shexp.weight q4_0 -blk.7.attn_v.weight q4_K +blk.8.attn_v.weight q4_K blk.8.ffn_down_exps.weight q4_0 blk.8.ffn_down_shexp.weight q4_0 -blk.8.attn_v.weight q4_K +blk.9.attn_v.weight q4_K blk.9.ffn_down_exps.weight q4_0 blk.9.ffn_down_shexp.weight q4_0 -blk.9.attn_v.weight q4_K +blk.10.attn_v.weight q4_K blk.10.ffn_down_exps.weight q4_0 blk.10.ffn_down_shexp.weight q4_0 -blk.10.attn_v.weight q4_K +blk.11.attn_v.weight q4_K blk.11.ffn_down_exps.weight q4_0 blk.11.ffn_down_shexp.weight q4_0 -blk.11.attn_v.weight q4_K +blk.12.attn_v.weight q4_K blk.12.ffn_down_exps.weight q4_0 blk.12.ffn_down_shexp.weight q4_0 -blk.12.attn_v.weight q4_K +blk.13.attn_v.weight q4_K blk.13.ffn_down_exps.weight q4_0 blk.13.ffn_down_shexp.weight q4_0 -blk.13.attn_v.weight q4_K +blk.14.attn_v.weight q4_K blk.14.ffn_down_exps.weight q4_0 blk.14.ffn_down_shexp.weight q4_0 -blk.14.attn_v.weight q4_K +blk.15.attn_v.weight q4_K blk.15.ffn_down_exps.weight q4_0 blk.15.ffn_down_shexp.weight q4_0 -blk.15.attn_v.weight q4_K +blk.16.attn_v.weight q4_K blk.16.ffn_down_exps.weight q4_0 blk.16.ffn_down_shexp.weight q4_0 -blk.16.attn_v.weight q4_K +blk.17.attn_v.weight q4_K blk.17.ffn_down_exps.weight q4_0 blk.17.ffn_down_shexp.weight q4_0 -blk.17.attn_v.weight q4_K +blk.18.attn_v.weight q4_K blk.18.ffn_down_exps.weight q4_0 blk.18.ffn_down_shexp.weight q4_0 -blk.18.attn_v.weight q4_K +blk.19.attn_v.weight q4_K blk.19.ffn_down_exps.weight q4_0 blk.19.ffn_down_shexp.weight q4_0 -blk.19.attn_v.weight q4_K +blk.20.attn_v.weight q4_K blk.20.ffn_down_exps.weight q4_0 blk.20.ffn_down_shexp.weight q4_0 -blk.20.attn_v.weight q4_K +blk.21.attn_v.weight q4_K blk.21.ffn_down_exps.weight q4_0 blk.21.ffn_down_shexp.weight q4_0 -blk.21.attn_v.weight q4_K +blk.22.attn_v.weight q4_K blk.22.ffn_down_exps.weight q4_0 blk.22.ffn_down_shexp.weight q4_0 -blk.22.attn_v.weight q4_K +blk.23.attn_v.weight q4_K blk.23.ffn_down_exps.weight q4_0 blk.23.ffn_down_shexp.weight q4_0 -blk.23.attn_v.weight q4_K +blk.24.attn_v.weight q4_K blk.24.ffn_down_exps.weight q4_0 blk.24.ffn_down_shexp.weight q4_0 -blk.24.attn_v.weight q4_K +blk.25.attn_v.weight q4_K blk.25.ffn_down_exps.weight q4_0 blk.25.ffn_down_shexp.weight q4_0 -blk.25.attn_v.weight q4_K +blk.26.attn_v.weight q4_K blk.26.ffn_down_exps.weight q4_0 blk.26.ffn_down_shexp.weight q4_0 -blk.26.attn_v.weight q4_K +blk.27.attn_v.weight q4_K blk.27.ffn_down_exps.weight q4_0 blk.27.ffn_down_shexp.weight q4_0 -blk.27.attn_v.weight q4_K +blk.28.attn_v.weight q4_K blk.28.ffn_down_exps.weight q4_0 blk.28.ffn_down_shexp.weight q4_0 -blk.28.attn_v.weight q4_K +blk.29.attn_v.weight q4_K blk.29.ffn_down_exps.weight q4_0 blk.29.ffn_down_shexp.weight q4_0 -blk.29.attn_v.weight q4_K +blk.30.attn_v.weight q4_K blk.30.ffn_down_exps.weight q4_0 blk.30.ffn_down_shexp.weight q4_0 -blk.30.attn_v.weight q4_K +blk.31.attn_v.weight q4_K blk.31.ffn_down_exps.weight q4_0 blk.31.ffn_down_shexp.weight q4_0 -blk.31.attn_v.weight q4_K +blk.32.attn_v.weight q4_K blk.32.ffn_down_exps.weight q4_0 blk.32.ffn_down_shexp.weight q4_0 -blk.32.attn_v.weight q4_K +blk.33.attn_v.weight q4_K blk.33.ffn_down_exps.weight q4_0 blk.33.ffn_down_shexp.weight q4_0 -blk.33.attn_v.weight q4_K +blk.34.attn_v.weight q4_K blk.34.ffn_down_exps.weight q4_0 blk.34.ffn_down_shexp.weight q4_0 -blk.34.attn_v.weight q4_K +blk.35.attn_v.weight q4_K blk.35.ffn_down_exps.weight q4_0 blk.35.ffn_down_shexp.weight q4_0 -blk.35.attn_v.weight q4_K +blk.36.attn_v.weight q4_K blk.36.ffn_down_exps.weight q4_0 blk.36.ffn_down_shexp.weight q4_0 -blk.36.attn_v.weight q4_K +blk.37.attn_v.weight q4_K blk.37.ffn_down_exps.weight q4_0 blk.37.ffn_down_shexp.weight q4_0 -blk.37.attn_v.weight q4_K +blk.38.attn_v.weight q4_K blk.38.ffn_down_exps.weight q4_0 blk.38.ffn_down_shexp.weight q4_0 -blk.38.attn_v.weight q4_K +blk.39.attn_v.weight q4_K blk.39.ffn_down_exps.weight q4_0 blk.39.ffn_down_shexp.weight q4_0 -blk.39.attn_v.weight q4_K +blk.40.attn_v.weight q4_K blk.40.ffn_down_exps.weight q4_0 blk.40.ffn_down_shexp.weight q4_0 -blk.40.attn_v.weight q4_K +blk.41.attn_v.weight q4_K blk.41.ffn_down_exps.weight q4_0 blk.41.ffn_down_shexp.weight q4_0 -blk.41.attn_v.weight q4_K +blk.42.attn_v.weight q4_K blk.42.ffn_down_exps.weight q4_0 blk.42.ffn_down_shexp.weight q4_0 -blk.42.attn_v.weight q4_K +blk.43.attn_v.weight q4_K blk.43.ffn_down_exps.weight q4_0 blk.43.ffn_down_shexp.weight q4_0 -blk.43.attn_v.weight q4_K +blk.44.attn_v.weight q4_K blk.44.ffn_down_exps.weight q4_0 blk.44.ffn_down_shexp.weight q4_0 -blk.44.attn_v.weight q4_K -output.weight q6_K +blk.45.attn_v.weight q4_K blk.45.ffn_down_exps.weight q4_0 blk.45.ffn_down_shexp.weight q4_0 -blk.45.attn_v.weight q4_K [IQ3_XS] iq3_s -blk.0.ffn_down.weight iq4_nl +output.weight q6_K blk.0.attn_k.weight iq3_xxs blk.0.attn_q.weight iq3_xxs blk.0.attn_v.weight q4_K -blk.1.ffn_down_exps.weight iq4_nl -blk.1.ffn_down_shexp.weight iq4_nl +blk.0.ffn_down.weight iq4_nl blk.1.attn_k.weight iq3_xxs blk.1.attn_q.weight iq3_xxs blk.1.attn_v.weight q4_K -blk.2.ffn_down_exps.weight iq4_nl -blk.2.ffn_down_shexp.weight iq4_nl +blk.1.ffn_down_exps.weight iq4_nl +blk.1.ffn_down_shexp.weight iq4_nl blk.2.attn_k.weight iq3_xxs blk.2.attn_q.weight iq3_xxs blk.2.attn_v.weight q4_K -blk.3.ffn_down_exps.weight iq4_nl -blk.3.ffn_down_shexp.weight iq4_nl +blk.2.ffn_down_exps.weight iq4_nl +blk.2.ffn_down_shexp.weight iq4_nl blk.3.attn_k.weight iq3_xxs blk.3.attn_q.weight iq3_xxs blk.3.attn_v.weight q4_K -blk.4.ffn_down_exps.weight iq4_nl -blk.4.ffn_down_shexp.weight iq4_nl +blk.3.ffn_down_exps.weight iq4_nl +blk.3.ffn_down_shexp.weight iq4_nl blk.4.attn_k.weight iq3_xxs blk.4.attn_q.weight iq3_xxs blk.4.attn_v.weight q4_K -blk.5.ffn_down_exps.weight iq4_nl -blk.5.ffn_gate_exps.weight iq3_xxs -blk.5.ffn_up_exps.weight iq3_xxs -blk.5.ffn_down_shexp.weight iq4_nl -blk.5.ffn_gate_shexp.weight iq3_xxs -blk.5.ffn_up_shexp.weight iq3_xxs +blk.4.ffn_down_exps.weight iq4_nl +blk.4.ffn_down_shexp.weight iq4_nl blk.5.attn_k.weight iq3_xxs blk.5.attn_q.weight iq3_xxs blk.5.attn_v.weight q4_K -blk.6.ffn_down_exps.weight iq4_nl -blk.6.ffn_gate_exps.weight iq3_xxs -blk.6.ffn_up_exps.weight iq3_xxs -blk.6.ffn_down_shexp.weight iq4_nl -blk.6.ffn_gate_shexp.weight iq3_xxs -blk.6.ffn_up_shexp.weight iq3_xxs +blk.5.ffn_down_exps.weight iq4_nl +blk.5.ffn_down_shexp.weight iq4_nl +blk.5.ffn_gate_exps.weight iq3_xxs +blk.5.ffn_gate_shexp.weight iq3_xxs +blk.5.ffn_up_exps.weight iq3_xxs +blk.5.ffn_up_shexp.weight iq3_xxs blk.6.attn_k.weight iq3_xxs blk.6.attn_q.weight iq3_xxs blk.6.attn_v.weight q4_K -blk.7.ffn_down_exps.weight iq4_nl -blk.7.ffn_gate_exps.weight iq3_xxs -blk.7.ffn_up_exps.weight iq3_xxs -blk.7.ffn_down_shexp.weight iq4_nl -blk.7.ffn_gate_shexp.weight iq3_xxs -blk.7.ffn_up_shexp.weight iq3_xxs +blk.6.ffn_down_exps.weight iq4_nl +blk.6.ffn_down_shexp.weight iq4_nl +blk.6.ffn_gate_exps.weight iq3_xxs +blk.6.ffn_gate_shexp.weight iq3_xxs +blk.6.ffn_up_exps.weight iq3_xxs +blk.6.ffn_up_shexp.weight iq3_xxs blk.7.attn_k.weight iq3_xxs blk.7.attn_q.weight iq3_xxs blk.7.attn_v.weight q4_K -blk.8.ffn_down_exps.weight iq4_nl -blk.8.ffn_gate_exps.weight iq3_xxs -blk.8.ffn_up_exps.weight iq3_xxs -blk.8.ffn_down_shexp.weight iq4_nl -blk.8.ffn_gate_shexp.weight iq3_xxs -blk.8.ffn_up_shexp.weight iq3_xxs +blk.7.ffn_down_exps.weight iq4_nl +blk.7.ffn_down_shexp.weight iq4_nl +blk.7.ffn_gate_exps.weight iq3_xxs +blk.7.ffn_gate_shexp.weight iq3_xxs +blk.7.ffn_up_exps.weight iq3_xxs +blk.7.ffn_up_shexp.weight iq3_xxs blk.8.attn_k.weight iq3_xxs blk.8.attn_q.weight iq3_xxs blk.8.attn_v.weight q4_K -blk.9.ffn_down_exps.weight iq4_nl -blk.9.ffn_gate_exps.weight iq3_xxs -blk.9.ffn_up_exps.weight iq3_xxs -blk.9.ffn_down_shexp.weight iq4_nl -blk.9.ffn_gate_shexp.weight iq3_xxs -blk.9.ffn_up_shexp.weight iq3_xxs +blk.8.ffn_down_exps.weight iq4_nl +blk.8.ffn_down_shexp.weight iq4_nl +blk.8.ffn_gate_exps.weight iq3_xxs +blk.8.ffn_gate_shexp.weight iq3_xxs +blk.8.ffn_up_exps.weight iq3_xxs +blk.8.ffn_up_shexp.weight iq3_xxs blk.9.attn_k.weight iq3_xxs blk.9.attn_q.weight iq3_xxs blk.9.attn_v.weight q4_K -blk.10.ffn_down_exps.weight iq4_nl -blk.10.ffn_gate_exps.weight iq3_xxs -blk.10.ffn_up_exps.weight iq3_xxs -blk.10.ffn_down_shexp.weight iq4_nl -blk.10.ffn_gate_shexp.weight iq3_xxs -blk.10.ffn_up_shexp.weight iq3_xxs +blk.9.ffn_down_exps.weight iq4_nl +blk.9.ffn_down_shexp.weight iq4_nl +blk.9.ffn_gate_exps.weight iq3_xxs +blk.9.ffn_gate_shexp.weight iq3_xxs +blk.9.ffn_up_exps.weight iq3_xxs +blk.9.ffn_up_shexp.weight iq3_xxs blk.10.attn_k.weight iq3_xxs blk.10.attn_q.weight iq3_xxs blk.10.attn_v.weight q4_K -blk.11.ffn_down_exps.weight iq4_nl -blk.11.ffn_gate_exps.weight iq3_xxs -blk.11.ffn_up_exps.weight iq3_xxs -blk.11.ffn_down_shexp.weight iq4_nl -blk.11.ffn_gate_shexp.weight iq3_xxs -blk.11.ffn_up_shexp.weight iq3_xxs +blk.10.ffn_down_exps.weight iq4_nl +blk.10.ffn_down_shexp.weight iq4_nl +blk.10.ffn_gate_exps.weight iq3_xxs +blk.10.ffn_gate_shexp.weight iq3_xxs +blk.10.ffn_up_exps.weight iq3_xxs +blk.10.ffn_up_shexp.weight iq3_xxs blk.11.attn_k.weight iq3_xxs blk.11.attn_q.weight iq3_xxs blk.11.attn_v.weight q4_K -blk.12.ffn_down_exps.weight iq4_nl -blk.12.ffn_gate_exps.weight iq3_xxs -blk.12.ffn_up_exps.weight iq3_xxs -blk.12.ffn_down_shexp.weight iq4_nl -blk.12.ffn_gate_shexp.weight iq3_xxs -blk.12.ffn_up_shexp.weight iq3_xxs +blk.11.ffn_down_exps.weight iq4_nl +blk.11.ffn_down_shexp.weight iq4_nl +blk.11.ffn_gate_exps.weight iq3_xxs +blk.11.ffn_gate_shexp.weight iq3_xxs +blk.11.ffn_up_exps.weight iq3_xxs +blk.11.ffn_up_shexp.weight iq3_xxs blk.12.attn_k.weight iq3_xxs blk.12.attn_q.weight iq3_xxs blk.12.attn_v.weight q4_K -blk.13.ffn_down_exps.weight iq4_nl -blk.13.ffn_gate_exps.weight iq3_xxs -blk.13.ffn_up_exps.weight iq3_xxs -blk.13.ffn_down_shexp.weight iq4_nl -blk.13.ffn_gate_shexp.weight iq3_xxs -blk.13.ffn_up_shexp.weight iq3_xxs +blk.12.ffn_down_exps.weight iq4_nl +blk.12.ffn_down_shexp.weight iq4_nl +blk.12.ffn_gate_exps.weight iq3_xxs +blk.12.ffn_gate_shexp.weight iq3_xxs +blk.12.ffn_up_exps.weight iq3_xxs +blk.12.ffn_up_shexp.weight iq3_xxs blk.13.attn_k.weight iq3_xxs blk.13.attn_q.weight iq3_xxs blk.13.attn_v.weight q4_K -blk.14.ffn_down_exps.weight iq4_nl -blk.14.ffn_gate_exps.weight iq3_xxs -blk.14.ffn_up_exps.weight iq3_xxs -blk.14.ffn_down_shexp.weight iq4_nl -blk.14.ffn_gate_shexp.weight iq3_xxs -blk.14.ffn_up_shexp.weight iq3_xxs +blk.13.ffn_down_exps.weight iq4_nl +blk.13.ffn_down_shexp.weight iq4_nl +blk.13.ffn_gate_exps.weight iq3_xxs +blk.13.ffn_gate_shexp.weight iq3_xxs +blk.13.ffn_up_exps.weight iq3_xxs +blk.13.ffn_up_shexp.weight iq3_xxs blk.14.attn_k.weight iq3_xxs blk.14.attn_q.weight iq3_xxs blk.14.attn_v.weight q4_K -blk.15.ffn_down_exps.weight iq4_nl -blk.15.ffn_gate_exps.weight iq3_xxs -blk.15.ffn_up_exps.weight iq3_xxs -blk.15.ffn_down_shexp.weight iq4_nl -blk.15.ffn_gate_shexp.weight iq3_xxs -blk.15.ffn_up_shexp.weight iq3_xxs +blk.14.ffn_down_exps.weight iq4_nl +blk.14.ffn_down_shexp.weight iq4_nl +blk.14.ffn_gate_exps.weight iq3_xxs +blk.14.ffn_gate_shexp.weight iq3_xxs +blk.14.ffn_up_exps.weight iq3_xxs +blk.14.ffn_up_shexp.weight iq3_xxs blk.15.attn_k.weight iq3_xxs blk.15.attn_q.weight iq3_xxs blk.15.attn_v.weight q4_K -blk.16.ffn_down_exps.weight iq4_nl -blk.16.ffn_gate_exps.weight iq3_xxs -blk.16.ffn_up_exps.weight iq3_xxs -blk.16.ffn_down_shexp.weight iq4_nl -blk.16.ffn_gate_shexp.weight iq3_xxs -blk.16.ffn_up_shexp.weight iq3_xxs +blk.15.ffn_down_exps.weight iq4_nl +blk.15.ffn_down_shexp.weight iq4_nl +blk.15.ffn_gate_exps.weight iq3_xxs +blk.15.ffn_gate_shexp.weight iq3_xxs +blk.15.ffn_up_exps.weight iq3_xxs +blk.15.ffn_up_shexp.weight iq3_xxs blk.16.attn_k.weight iq3_xxs blk.16.attn_q.weight iq3_xxs blk.16.attn_v.weight q4_K -blk.17.ffn_down_exps.weight iq4_nl -blk.17.ffn_gate_exps.weight iq3_xxs -blk.17.ffn_up_exps.weight iq3_xxs -blk.17.ffn_down_shexp.weight iq4_nl -blk.17.ffn_gate_shexp.weight iq3_xxs -blk.17.ffn_up_shexp.weight iq3_xxs +blk.16.ffn_down_exps.weight iq4_nl +blk.16.ffn_down_shexp.weight iq4_nl +blk.16.ffn_gate_exps.weight iq3_xxs +blk.16.ffn_gate_shexp.weight iq3_xxs +blk.16.ffn_up_exps.weight iq3_xxs +blk.16.ffn_up_shexp.weight iq3_xxs blk.17.attn_k.weight iq3_xxs blk.17.attn_q.weight iq3_xxs blk.17.attn_v.weight q4_K -blk.18.ffn_down_exps.weight iq4_nl -blk.18.ffn_gate_exps.weight iq3_xxs -blk.18.ffn_up_exps.weight iq3_xxs -blk.18.ffn_down_shexp.weight iq4_nl -blk.18.ffn_gate_shexp.weight iq3_xxs -blk.18.ffn_up_shexp.weight iq3_xxs +blk.17.ffn_down_exps.weight iq4_nl +blk.17.ffn_down_shexp.weight iq4_nl +blk.17.ffn_gate_exps.weight iq3_xxs +blk.17.ffn_gate_shexp.weight iq3_xxs +blk.17.ffn_up_exps.weight iq3_xxs +blk.17.ffn_up_shexp.weight iq3_xxs blk.18.attn_k.weight iq3_xxs blk.18.attn_q.weight iq3_xxs blk.18.attn_v.weight q4_K -blk.19.ffn_down_exps.weight iq4_nl -blk.19.ffn_gate_exps.weight iq3_xxs -blk.19.ffn_up_exps.weight iq3_xxs -blk.19.ffn_down_shexp.weight iq4_nl -blk.19.ffn_gate_shexp.weight iq3_xxs -blk.19.ffn_up_shexp.weight iq3_xxs +blk.18.ffn_down_exps.weight iq4_nl +blk.18.ffn_down_shexp.weight iq4_nl +blk.18.ffn_gate_exps.weight iq3_xxs +blk.18.ffn_gate_shexp.weight iq3_xxs +blk.18.ffn_up_exps.weight iq3_xxs +blk.18.ffn_up_shexp.weight iq3_xxs blk.19.attn_k.weight iq3_xxs blk.19.attn_q.weight iq3_xxs blk.19.attn_v.weight q4_K -blk.20.ffn_down_exps.weight iq4_nl -blk.20.ffn_gate_exps.weight iq3_xxs -blk.20.ffn_up_exps.weight iq3_xxs -blk.20.ffn_down_shexp.weight iq4_nl -blk.20.ffn_gate_shexp.weight iq3_xxs -blk.20.ffn_up_shexp.weight iq3_xxs +blk.19.ffn_down_exps.weight iq4_nl +blk.19.ffn_down_shexp.weight iq4_nl +blk.19.ffn_gate_exps.weight iq3_xxs +blk.19.ffn_gate_shexp.weight iq3_xxs +blk.19.ffn_up_exps.weight iq3_xxs +blk.19.ffn_up_shexp.weight iq3_xxs blk.20.attn_k.weight iq3_xxs blk.20.attn_q.weight iq3_xxs blk.20.attn_v.weight q4_K -blk.21.ffn_down_exps.weight iq4_nl -blk.21.ffn_gate_exps.weight iq3_xxs -blk.21.ffn_up_exps.weight iq3_xxs -blk.21.ffn_down_shexp.weight iq4_nl -blk.21.ffn_gate_shexp.weight iq3_xxs -blk.21.ffn_up_shexp.weight iq3_xxs +blk.20.ffn_down_exps.weight iq4_nl +blk.20.ffn_down_shexp.weight iq4_nl +blk.20.ffn_gate_exps.weight iq3_xxs +blk.20.ffn_gate_shexp.weight iq3_xxs +blk.20.ffn_up_exps.weight iq3_xxs +blk.20.ffn_up_shexp.weight iq3_xxs blk.21.attn_k.weight iq3_xxs blk.21.attn_q.weight iq3_xxs blk.21.attn_v.weight q4_K -blk.22.ffn_down_exps.weight iq4_nl -blk.22.ffn_gate_exps.weight iq3_xxs -blk.22.ffn_up_exps.weight iq3_xxs -blk.22.ffn_down_shexp.weight iq4_nl -blk.22.ffn_gate_shexp.weight iq3_xxs -blk.22.ffn_up_shexp.weight iq3_xxs +blk.21.ffn_down_exps.weight iq4_nl +blk.21.ffn_down_shexp.weight iq4_nl +blk.21.ffn_gate_exps.weight iq3_xxs +blk.21.ffn_gate_shexp.weight iq3_xxs +blk.21.ffn_up_exps.weight iq3_xxs +blk.21.ffn_up_shexp.weight iq3_xxs blk.22.attn_k.weight iq3_xxs blk.22.attn_q.weight iq3_xxs blk.22.attn_v.weight q4_K -blk.23.ffn_down_exps.weight iq4_nl -blk.23.ffn_gate_exps.weight iq3_xxs -blk.23.ffn_up_exps.weight iq3_xxs -blk.23.ffn_down_shexp.weight iq4_nl -blk.23.ffn_gate_shexp.weight iq3_xxs -blk.23.ffn_up_shexp.weight iq3_xxs +blk.22.ffn_down_exps.weight iq4_nl +blk.22.ffn_down_shexp.weight iq4_nl +blk.22.ffn_gate_exps.weight iq3_xxs +blk.22.ffn_gate_shexp.weight iq3_xxs +blk.22.ffn_up_exps.weight iq3_xxs +blk.22.ffn_up_shexp.weight iq3_xxs blk.23.attn_k.weight iq3_xxs blk.23.attn_q.weight iq3_xxs blk.23.attn_v.weight q4_K -blk.24.ffn_down_exps.weight iq4_nl -blk.24.ffn_gate_exps.weight iq3_xxs -blk.24.ffn_up_exps.weight iq3_xxs -blk.24.ffn_down_shexp.weight iq4_nl -blk.24.ffn_gate_shexp.weight iq3_xxs -blk.24.ffn_up_shexp.weight iq3_xxs +blk.23.ffn_down_exps.weight iq4_nl +blk.23.ffn_down_shexp.weight iq4_nl +blk.23.ffn_gate_exps.weight iq3_xxs +blk.23.ffn_gate_shexp.weight iq3_xxs +blk.23.ffn_up_exps.weight iq3_xxs +blk.23.ffn_up_shexp.weight iq3_xxs blk.24.attn_k.weight iq3_xxs blk.24.attn_q.weight iq3_xxs blk.24.attn_v.weight q4_K -blk.25.ffn_down_exps.weight iq4_nl -blk.25.ffn_gate_exps.weight iq3_xxs -blk.25.ffn_up_exps.weight iq3_xxs -blk.25.ffn_down_shexp.weight iq4_nl -blk.25.ffn_gate_shexp.weight iq3_xxs -blk.25.ffn_up_shexp.weight iq3_xxs +blk.24.ffn_down_exps.weight iq4_nl +blk.24.ffn_down_shexp.weight iq4_nl +blk.24.ffn_gate_exps.weight iq3_xxs +blk.24.ffn_gate_shexp.weight iq3_xxs +blk.24.ffn_up_exps.weight iq3_xxs +blk.24.ffn_up_shexp.weight iq3_xxs blk.25.attn_k.weight iq3_xxs blk.25.attn_q.weight iq3_xxs blk.25.attn_v.weight q4_K -blk.26.ffn_down_exps.weight iq4_nl -blk.26.ffn_gate_exps.weight iq3_xxs -blk.26.ffn_up_exps.weight iq3_xxs -blk.26.ffn_down_shexp.weight iq4_nl -blk.26.ffn_gate_shexp.weight iq3_xxs -blk.26.ffn_up_shexp.weight iq3_xxs +blk.25.ffn_down_exps.weight iq4_nl +blk.25.ffn_down_shexp.weight iq4_nl +blk.25.ffn_gate_exps.weight iq3_xxs +blk.25.ffn_gate_shexp.weight iq3_xxs +blk.25.ffn_up_exps.weight iq3_xxs +blk.25.ffn_up_shexp.weight iq3_xxs blk.26.attn_k.weight iq3_xxs blk.26.attn_q.weight iq3_xxs blk.26.attn_v.weight q4_K -blk.27.ffn_down_exps.weight iq4_nl -blk.27.ffn_gate_exps.weight iq3_xxs -blk.27.ffn_up_exps.weight iq3_xxs -blk.27.ffn_down_shexp.weight iq4_nl -blk.27.ffn_gate_shexp.weight iq3_xxs -blk.27.ffn_up_shexp.weight iq3_xxs +blk.26.ffn_down_exps.weight iq4_nl +blk.26.ffn_down_shexp.weight iq4_nl +blk.26.ffn_gate_exps.weight iq3_xxs +blk.26.ffn_gate_shexp.weight iq3_xxs +blk.26.ffn_up_exps.weight iq3_xxs +blk.26.ffn_up_shexp.weight iq3_xxs blk.27.attn_k.weight iq3_xxs blk.27.attn_q.weight iq3_xxs blk.27.attn_v.weight q4_K -blk.28.ffn_down_exps.weight iq4_nl -blk.28.ffn_gate_exps.weight iq3_xxs -blk.28.ffn_up_exps.weight iq3_xxs -blk.28.ffn_down_shexp.weight iq4_nl -blk.28.ffn_gate_shexp.weight iq3_xxs -blk.28.ffn_up_shexp.weight iq3_xxs +blk.27.ffn_down_exps.weight iq4_nl +blk.27.ffn_down_shexp.weight iq4_nl +blk.27.ffn_gate_exps.weight iq3_xxs +blk.27.ffn_gate_shexp.weight iq3_xxs +blk.27.ffn_up_exps.weight iq3_xxs +blk.27.ffn_up_shexp.weight iq3_xxs blk.28.attn_k.weight iq3_xxs blk.28.attn_q.weight iq3_xxs blk.28.attn_v.weight q4_K -blk.29.ffn_down_exps.weight iq4_nl -blk.29.ffn_gate_exps.weight iq3_xxs -blk.29.ffn_up_exps.weight iq3_xxs -blk.29.ffn_down_shexp.weight iq4_nl -blk.29.ffn_gate_shexp.weight iq3_xxs -blk.29.ffn_up_shexp.weight iq3_xxs +blk.28.ffn_down_exps.weight iq4_nl +blk.28.ffn_down_shexp.weight iq4_nl +blk.28.ffn_gate_exps.weight iq3_xxs +blk.28.ffn_gate_shexp.weight iq3_xxs +blk.28.ffn_up_exps.weight iq3_xxs +blk.28.ffn_up_shexp.weight iq3_xxs blk.29.attn_k.weight iq3_xxs blk.29.attn_q.weight iq3_xxs blk.29.attn_v.weight q4_K -blk.30.ffn_down_exps.weight iq4_nl -blk.30.ffn_gate_exps.weight iq3_xxs -blk.30.ffn_up_exps.weight iq3_xxs -blk.30.ffn_down_shexp.weight iq4_nl -blk.30.ffn_gate_shexp.weight iq3_xxs -blk.30.ffn_up_shexp.weight iq3_xxs +blk.29.ffn_down_exps.weight iq4_nl +blk.29.ffn_down_shexp.weight iq4_nl +blk.29.ffn_gate_exps.weight iq3_xxs +blk.29.ffn_gate_shexp.weight iq3_xxs +blk.29.ffn_up_exps.weight iq3_xxs +blk.29.ffn_up_shexp.weight iq3_xxs blk.30.attn_k.weight iq3_xxs blk.30.attn_q.weight iq3_xxs blk.30.attn_v.weight q4_K -blk.31.ffn_down_exps.weight iq4_nl -blk.31.ffn_gate_exps.weight iq3_xxs -blk.31.ffn_up_exps.weight iq3_xxs -blk.31.ffn_down_shexp.weight iq4_nl -blk.31.ffn_gate_shexp.weight iq3_xxs -blk.31.ffn_up_shexp.weight iq3_xxs +blk.30.ffn_down_exps.weight iq4_nl +blk.30.ffn_down_shexp.weight iq4_nl +blk.30.ffn_gate_exps.weight iq3_xxs +blk.30.ffn_gate_shexp.weight iq3_xxs +blk.30.ffn_up_exps.weight iq3_xxs +blk.30.ffn_up_shexp.weight iq3_xxs blk.31.attn_k.weight iq3_xxs blk.31.attn_q.weight iq3_xxs blk.31.attn_v.weight q4_K -blk.32.ffn_down_exps.weight iq4_nl -blk.32.ffn_gate_exps.weight iq3_xxs -blk.32.ffn_up_exps.weight iq3_xxs -blk.32.ffn_down_shexp.weight iq4_nl -blk.32.ffn_gate_shexp.weight iq3_xxs -blk.32.ffn_up_shexp.weight iq3_xxs +blk.31.ffn_down_exps.weight iq4_nl +blk.31.ffn_down_shexp.weight iq4_nl +blk.31.ffn_gate_exps.weight iq3_xxs +blk.31.ffn_gate_shexp.weight iq3_xxs +blk.31.ffn_up_exps.weight iq3_xxs +blk.31.ffn_up_shexp.weight iq3_xxs blk.32.attn_k.weight iq3_xxs blk.32.attn_q.weight iq3_xxs blk.32.attn_v.weight q4_K -blk.33.ffn_down_exps.weight iq4_nl -blk.33.ffn_gate_exps.weight iq3_xxs -blk.33.ffn_up_exps.weight iq3_xxs -blk.33.ffn_down_shexp.weight iq4_nl -blk.33.ffn_gate_shexp.weight iq3_xxs -blk.33.ffn_up_shexp.weight iq3_xxs +blk.32.ffn_down_exps.weight iq4_nl +blk.32.ffn_down_shexp.weight iq4_nl +blk.32.ffn_gate_exps.weight iq3_xxs +blk.32.ffn_gate_shexp.weight iq3_xxs +blk.32.ffn_up_exps.weight iq3_xxs +blk.32.ffn_up_shexp.weight iq3_xxs blk.33.attn_k.weight iq3_xxs blk.33.attn_q.weight iq3_xxs blk.33.attn_v.weight q4_K -blk.34.ffn_down_exps.weight iq4_nl -blk.34.ffn_gate_exps.weight iq3_xxs -blk.34.ffn_up_exps.weight iq3_xxs -blk.34.ffn_gate_shexp.weight iq3_xxs -blk.34.ffn_down_shexp.weight iq4_nl -blk.34.ffn_up_shexp.weight iq3_xxs +blk.33.ffn_down_exps.weight iq4_nl +blk.33.ffn_down_shexp.weight iq4_nl +blk.33.ffn_gate_exps.weight iq3_xxs +blk.33.ffn_gate_shexp.weight iq3_xxs +blk.33.ffn_up_exps.weight iq3_xxs +blk.33.ffn_up_shexp.weight iq3_xxs blk.34.attn_k.weight iq3_xxs blk.34.attn_q.weight iq3_xxs blk.34.attn_v.weight q4_K -blk.35.ffn_down_exps.weight iq4_nl -blk.35.ffn_gate_exps.weight iq3_xxs -blk.35.ffn_up_exps.weight iq3_xxs -blk.35.ffn_down_shexp.weight iq4_nl -blk.35.ffn_gate_shexp.weight iq3_xxs -blk.35.ffn_up_shexp.weight iq3_xxs +blk.34.ffn_down_exps.weight iq4_nl +blk.34.ffn_down_shexp.weight iq4_nl +blk.34.ffn_gate_exps.weight iq3_xxs +blk.34.ffn_gate_shexp.weight iq3_xxs +blk.34.ffn_up_exps.weight iq3_xxs +blk.34.ffn_up_shexp.weight iq3_xxs blk.35.attn_k.weight iq3_xxs blk.35.attn_q.weight iq3_xxs blk.35.attn_v.weight q4_K -blk.36.ffn_down_exps.weight iq4_nl -blk.36.ffn_gate_exps.weight iq3_xxs -blk.36.ffn_up_exps.weight iq3_xxs -blk.36.ffn_down_shexp.weight iq4_nl -blk.36.ffn_gate_shexp.weight iq3_xxs -blk.36.ffn_up_shexp.weight iq3_xxs +blk.35.ffn_down_exps.weight iq4_nl +blk.35.ffn_down_shexp.weight iq4_nl +blk.35.ffn_gate_exps.weight iq3_xxs +blk.35.ffn_gate_shexp.weight iq3_xxs +blk.35.ffn_up_exps.weight iq3_xxs +blk.35.ffn_up_shexp.weight iq3_xxs blk.36.attn_k.weight iq3_xxs blk.36.attn_q.weight iq3_xxs blk.36.attn_v.weight q4_K -blk.37.ffn_down_exps.weight iq4_nl -blk.37.ffn_gate_exps.weight iq3_xxs -blk.37.ffn_up_exps.weight iq3_xxs -blk.37.ffn_down_shexp.weight iq4_nl -blk.37.ffn_gate_shexp.weight iq3_xxs -blk.37.ffn_up_shexp.weight iq3_xxs +blk.36.ffn_down_exps.weight iq4_nl +blk.36.ffn_down_shexp.weight iq4_nl +blk.36.ffn_gate_exps.weight iq3_xxs +blk.36.ffn_gate_shexp.weight iq3_xxs +blk.36.ffn_up_exps.weight iq3_xxs +blk.36.ffn_up_shexp.weight iq3_xxs blk.37.attn_k.weight iq3_xxs blk.37.attn_q.weight iq3_xxs blk.37.attn_v.weight q4_K -blk.38.ffn_down_exps.weight iq4_nl -blk.38.ffn_gate_exps.weight iq3_xxs -blk.38.ffn_up_exps.weight iq3_xxs -blk.38.ffn_down_shexp.weight iq4_nl -blk.38.ffn_gate_shexp.weight iq3_xxs -blk.38.ffn_up_shexp.weight iq3_xxs +blk.37.ffn_down_exps.weight iq4_nl +blk.37.ffn_down_shexp.weight iq4_nl +blk.37.ffn_gate_exps.weight iq3_xxs +blk.37.ffn_gate_shexp.weight iq3_xxs +blk.37.ffn_up_exps.weight iq3_xxs +blk.37.ffn_up_shexp.weight iq3_xxs blk.38.attn_k.weight iq3_xxs blk.38.attn_q.weight iq3_xxs blk.38.attn_v.weight q4_K -blk.39.ffn_down_exps.weight iq4_nl -blk.39.ffn_gate_exps.weight iq3_xxs -blk.39.ffn_up_exps.weight iq3_xxs -blk.39.ffn_down_shexp.weight iq4_nl -blk.39.ffn_gate_shexp.weight iq3_xxs -blk.39.ffn_up_shexp.weight iq3_xxs +blk.38.ffn_down_exps.weight iq4_nl +blk.38.ffn_down_shexp.weight iq4_nl +blk.38.ffn_gate_exps.weight iq3_xxs +blk.38.ffn_gate_shexp.weight iq3_xxs +blk.38.ffn_up_exps.weight iq3_xxs +blk.38.ffn_up_shexp.weight iq3_xxs blk.39.attn_k.weight iq3_xxs blk.39.attn_q.weight iq3_xxs blk.39.attn_v.weight q4_K -blk.40.ffn_down_exps.weight iq4_nl -blk.40.ffn_down_shexp.weight iq4_nl +blk.39.ffn_down_exps.weight iq4_nl +blk.39.ffn_down_shexp.weight iq4_nl +blk.39.ffn_gate_exps.weight iq3_xxs +blk.39.ffn_gate_shexp.weight iq3_xxs +blk.39.ffn_up_exps.weight iq3_xxs +blk.39.ffn_up_shexp.weight iq3_xxs blk.40.attn_k.weight iq3_xxs blk.40.attn_q.weight iq3_xxs blk.40.attn_v.weight q4_K -blk.41.ffn_down_exps.weight iq4_nl -blk.41.ffn_down_shexp.weight iq4_nl +blk.40.ffn_down_exps.weight iq4_nl +blk.40.ffn_down_shexp.weight iq4_nl blk.41.attn_k.weight iq3_xxs blk.41.attn_q.weight iq3_xxs blk.41.attn_v.weight q4_K -blk.42.ffn_down_exps.weight iq4_nl -blk.42.ffn_down_shexp.weight iq4_nl +blk.41.ffn_down_exps.weight iq4_nl +blk.41.ffn_down_shexp.weight iq4_nl blk.42.attn_k.weight iq3_xxs blk.42.attn_q.weight iq3_xxs blk.42.attn_v.weight q4_K -blk.43.ffn_down_exps.weight iq4_nl -blk.43.ffn_down_shexp.weight iq4_nl +blk.42.ffn_down_exps.weight iq4_nl +blk.42.ffn_down_shexp.weight iq4_nl blk.43.attn_k.weight iq3_xxs blk.43.attn_q.weight iq3_xxs blk.43.attn_v.weight q4_K -blk.44.ffn_down_exps.weight iq4_nl -blk.44.ffn_down_shexp.weight iq4_nl +blk.43.ffn_down_exps.weight iq4_nl +blk.43.ffn_down_shexp.weight iq4_nl blk.44.attn_k.weight iq3_xxs blk.44.attn_q.weight iq3_xxs blk.44.attn_v.weight q4_K -output.weight q6_K -blk.45.ffn_down_exps.weight iq4_nl -blk.45.ffn_down_shexp.weight iq4_nl +blk.44.ffn_down_exps.weight iq4_nl +blk.44.ffn_down_shexp.weight iq4_nl blk.45.attn_k.weight iq3_xxs blk.45.attn_q.weight iq3_xxs blk.45.attn_v.weight q4_K +blk.45.ffn_down_exps.weight iq4_nl +blk.45.ffn_down_shexp.weight iq4_nl [IQ3_XXS] iq3_xxs +output.weight q5_K token_embd.weight iq3_s -blk.0.ffn_down.weight q5_0 blk.0.attn_k.weight iq2_s blk.0.attn_output.weight iq3_s blk.0.attn_q.weight iq2_s blk.0.attn_v.weight q4_K -blk.1.ffn_down_exps.weight q5_0 -blk.1.ffn_down_shexp.weight q5_0 +blk.0.ffn_down.weight q5_0 blk.1.attn_k.weight iq2_s blk.1.attn_output.weight iq3_s blk.1.attn_q.weight iq2_s blk.1.attn_v.weight q4_K -blk.2.ffn_down_exps.weight q5_0 -blk.2.ffn_down_shexp.weight q5_0 +blk.1.ffn_down_exps.weight q5_0 +blk.1.ffn_down_shexp.weight q5_0 blk.2.attn_k.weight iq2_s blk.2.attn_output.weight iq3_s blk.2.attn_q.weight iq2_s blk.2.attn_v.weight q4_K -blk.3.ffn_down_exps.weight q5_0 -blk.3.ffn_down_shexp.weight q5_0 +blk.2.ffn_down_exps.weight q5_0 +blk.2.ffn_down_shexp.weight q5_0 blk.3.attn_k.weight iq2_s blk.3.attn_output.weight iq3_s blk.3.attn_q.weight iq2_s blk.3.attn_v.weight q4_K -blk.4.ffn_down_exps.weight q5_0 -blk.4.ffn_down_shexp.weight q5_0 +blk.3.ffn_down_exps.weight q5_0 +blk.3.ffn_down_shexp.weight q5_0 blk.4.attn_k.weight iq2_s blk.4.attn_output.weight iq3_s blk.4.attn_q.weight iq2_s blk.4.attn_v.weight q4_K -blk.5.ffn_down_exps.weight q4_0 -blk.5.ffn_down_shexp.weight q4_0 +blk.4.ffn_down_exps.weight q5_0 +blk.4.ffn_down_shexp.weight q5_0 blk.5.attn_k.weight iq2_s blk.5.attn_output.weight iq3_s blk.5.attn_q.weight iq2_s blk.5.attn_v.weight q4_K -blk.6.ffn_down_exps.weight q4_0 -blk.6.ffn_down_shexp.weight q4_0 +blk.5.ffn_down_exps.weight q4_0 +blk.5.ffn_down_shexp.weight q4_0 blk.6.attn_k.weight iq2_s blk.6.attn_output.weight iq3_s blk.6.attn_q.weight iq2_s blk.6.attn_v.weight q4_K -blk.7.ffn_down_exps.weight q4_0 -blk.7.ffn_down_shexp.weight q4_0 +blk.6.ffn_down_exps.weight q4_0 +blk.6.ffn_down_shexp.weight q4_0 blk.7.attn_k.weight iq2_s blk.7.attn_output.weight iq3_s blk.7.attn_q.weight iq2_s blk.7.attn_v.weight q4_K -blk.8.ffn_down_exps.weight q4_0 -blk.8.ffn_down_shexp.weight q4_0 +blk.7.ffn_down_exps.weight q4_0 +blk.7.ffn_down_shexp.weight q4_0 blk.8.attn_k.weight iq2_s blk.8.attn_output.weight iq3_s blk.8.attn_q.weight iq2_s blk.8.attn_v.weight q4_K -blk.9.ffn_down_exps.weight q4_0 -blk.9.ffn_down_shexp.weight q4_0 +blk.8.ffn_down_exps.weight q4_0 +blk.8.ffn_down_shexp.weight q4_0 blk.9.attn_k.weight iq2_s blk.9.attn_output.weight iq3_s blk.9.attn_q.weight iq2_s blk.9.attn_v.weight q4_K -blk.10.ffn_down_exps.weight q4_0 -blk.10.ffn_down_shexp.weight q4_0 +blk.9.ffn_down_exps.weight q4_0 +blk.9.ffn_down_shexp.weight q4_0 blk.10.attn_k.weight iq2_s blk.10.attn_output.weight iq3_s blk.10.attn_q.weight iq2_s blk.10.attn_v.weight q4_K -blk.11.ffn_down_exps.weight q4_0 -blk.11.ffn_down_shexp.weight q4_0 +blk.10.ffn_down_exps.weight q4_0 +blk.10.ffn_down_shexp.weight q4_0 blk.11.attn_k.weight iq2_s blk.11.attn_output.weight iq3_s blk.11.attn_q.weight iq2_s blk.11.attn_v.weight q4_K -blk.12.ffn_down_exps.weight q4_0 -blk.12.ffn_down_shexp.weight q4_0 +blk.11.ffn_down_exps.weight q4_0 +blk.11.ffn_down_shexp.weight q4_0 blk.12.attn_k.weight iq2_s blk.12.attn_output.weight iq3_s blk.12.attn_q.weight iq2_s blk.12.attn_v.weight q4_K -blk.13.ffn_down_exps.weight q4_0 -blk.13.ffn_down_shexp.weight q4_0 +blk.12.ffn_down_exps.weight q4_0 +blk.12.ffn_down_shexp.weight q4_0 blk.13.attn_k.weight iq2_s blk.13.attn_output.weight iq3_s blk.13.attn_q.weight iq2_s blk.13.attn_v.weight q4_K -blk.14.ffn_down_exps.weight q4_0 -blk.14.ffn_down_shexp.weight q4_0 +blk.13.ffn_down_exps.weight q4_0 +blk.13.ffn_down_shexp.weight q4_0 blk.14.attn_k.weight iq2_s blk.14.attn_output.weight iq3_s blk.14.attn_q.weight iq2_s blk.14.attn_v.weight q4_K -blk.15.ffn_down_exps.weight q4_0 -blk.15.ffn_down_shexp.weight q4_0 +blk.14.ffn_down_exps.weight q4_0 +blk.14.ffn_down_shexp.weight q4_0 blk.15.attn_k.weight iq2_s blk.15.attn_output.weight iq3_s blk.15.attn_q.weight iq2_s blk.15.attn_v.weight q4_K -blk.16.ffn_down_exps.weight q4_0 -blk.16.ffn_down_shexp.weight q4_0 +blk.15.ffn_down_exps.weight q4_0 +blk.15.ffn_down_shexp.weight q4_0 blk.16.attn_k.weight iq2_s blk.16.attn_output.weight iq3_s blk.16.attn_q.weight iq2_s blk.16.attn_v.weight q4_K -blk.17.ffn_down_exps.weight q4_0 -blk.17.ffn_down_shexp.weight q4_0 +blk.16.ffn_down_exps.weight q4_0 +blk.16.ffn_down_shexp.weight q4_0 blk.17.attn_k.weight iq2_s blk.17.attn_output.weight iq3_s blk.17.attn_q.weight iq2_s blk.17.attn_v.weight q4_K -blk.18.ffn_down_exps.weight q4_0 -blk.18.ffn_down_shexp.weight q4_0 +blk.17.ffn_down_exps.weight q4_0 +blk.17.ffn_down_shexp.weight q4_0 blk.18.attn_k.weight iq2_s blk.18.attn_output.weight iq3_s blk.18.attn_q.weight iq2_s blk.18.attn_v.weight q4_K -blk.19.ffn_down_exps.weight q4_0 -blk.19.ffn_down_shexp.weight q4_0 +blk.18.ffn_down_exps.weight q4_0 +blk.18.ffn_down_shexp.weight q4_0 blk.19.attn_k.weight iq2_s blk.19.attn_output.weight iq3_s blk.19.attn_q.weight iq2_s blk.19.attn_v.weight q4_K -blk.20.ffn_down_exps.weight q4_0 -blk.20.ffn_down_shexp.weight q4_0 +blk.19.ffn_down_exps.weight q4_0 +blk.19.ffn_down_shexp.weight q4_0 blk.20.attn_k.weight iq2_s blk.20.attn_output.weight iq3_s blk.20.attn_q.weight iq2_s blk.20.attn_v.weight q4_K -blk.21.ffn_down_exps.weight q4_0 -blk.21.ffn_down_shexp.weight q4_0 +blk.20.ffn_down_exps.weight q4_0 +blk.20.ffn_down_shexp.weight q4_0 blk.21.attn_k.weight iq2_s blk.21.attn_output.weight iq3_s blk.21.attn_q.weight iq2_s blk.21.attn_v.weight q4_K -blk.22.ffn_down_exps.weight q4_0 -blk.22.ffn_down_shexp.weight q4_0 +blk.21.ffn_down_exps.weight q4_0 +blk.21.ffn_down_shexp.weight q4_0 blk.22.attn_k.weight iq2_s blk.22.attn_output.weight iq3_s blk.22.attn_q.weight iq2_s blk.22.attn_v.weight q4_K -blk.23.ffn_down_exps.weight q4_0 -blk.23.ffn_down_shexp.weight q4_0 +blk.22.ffn_down_exps.weight q4_0 +blk.22.ffn_down_shexp.weight q4_0 blk.23.attn_k.weight iq2_s blk.23.attn_output.weight iq3_s blk.23.attn_q.weight iq2_s blk.23.attn_v.weight q4_K -blk.24.ffn_down_exps.weight q4_0 -blk.24.ffn_down_shexp.weight q4_0 +blk.23.ffn_down_exps.weight q4_0 +blk.23.ffn_down_shexp.weight q4_0 blk.24.attn_k.weight iq2_s blk.24.attn_output.weight iq3_s blk.24.attn_q.weight iq2_s blk.24.attn_v.weight q4_K -blk.25.ffn_down_exps.weight q4_0 -blk.25.ffn_down_shexp.weight q4_0 +blk.24.ffn_down_exps.weight q4_0 +blk.24.ffn_down_shexp.weight q4_0 blk.25.attn_k.weight iq2_s blk.25.attn_output.weight iq3_s blk.25.attn_q.weight iq2_s blk.25.attn_v.weight q4_K -blk.26.ffn_down_exps.weight q4_0 -blk.26.ffn_down_shexp.weight q4_0 +blk.25.ffn_down_exps.weight q4_0 +blk.25.ffn_down_shexp.weight q4_0 blk.26.attn_k.weight iq2_s blk.26.attn_output.weight iq3_s blk.26.attn_q.weight iq2_s blk.26.attn_v.weight q4_K -blk.27.ffn_down_exps.weight q4_0 -blk.27.ffn_down_shexp.weight q4_0 +blk.26.ffn_down_exps.weight q4_0 +blk.26.ffn_down_shexp.weight q4_0 blk.27.attn_k.weight iq2_s blk.27.attn_output.weight iq3_s blk.27.attn_q.weight iq2_s blk.27.attn_v.weight q4_K -blk.28.ffn_down_exps.weight q4_0 -blk.28.ffn_down_shexp.weight q4_0 +blk.27.ffn_down_exps.weight q4_0 +blk.27.ffn_down_shexp.weight q4_0 blk.28.attn_k.weight iq2_s blk.28.attn_output.weight iq3_s blk.28.attn_q.weight iq2_s blk.28.attn_v.weight q4_K -blk.29.ffn_down_exps.weight q4_0 -blk.29.ffn_down_shexp.weight q4_0 +blk.28.ffn_down_exps.weight q4_0 +blk.28.ffn_down_shexp.weight q4_0 blk.29.attn_k.weight iq2_s blk.29.attn_output.weight iq3_s blk.29.attn_q.weight iq2_s blk.29.attn_v.weight q4_K -blk.30.ffn_down_exps.weight q4_0 -blk.30.ffn_down_shexp.weight q4_0 +blk.29.ffn_down_exps.weight q4_0 +blk.29.ffn_down_shexp.weight q4_0 blk.30.attn_k.weight iq2_s blk.30.attn_output.weight iq3_s blk.30.attn_q.weight iq2_s blk.30.attn_v.weight q4_K -blk.31.ffn_down_exps.weight q4_0 -blk.31.ffn_down_shexp.weight q4_0 +blk.30.ffn_down_exps.weight q4_0 +blk.30.ffn_down_shexp.weight q4_0 blk.31.attn_k.weight iq2_s blk.31.attn_output.weight iq3_s blk.31.attn_q.weight iq2_s blk.31.attn_v.weight q4_K -blk.32.ffn_down_exps.weight q4_0 -blk.32.ffn_down_shexp.weight q4_0 +blk.31.ffn_down_exps.weight q4_0 +blk.31.ffn_down_shexp.weight q4_0 blk.32.attn_k.weight iq2_s blk.32.attn_output.weight iq3_s blk.32.attn_q.weight iq2_s blk.32.attn_v.weight q4_K -blk.33.ffn_down_exps.weight q4_0 -blk.33.ffn_down_shexp.weight q4_0 +blk.32.ffn_down_exps.weight q4_0 +blk.32.ffn_down_shexp.weight q4_0 blk.33.attn_k.weight iq2_s blk.33.attn_output.weight iq3_s blk.33.attn_q.weight iq2_s blk.33.attn_v.weight q4_K -blk.34.ffn_down_exps.weight q4_0 -blk.34.attn_output.weight iq3_s -blk.34.ffn_down_shexp.weight q4_0 +blk.33.ffn_down_exps.weight q4_0 +blk.33.ffn_down_shexp.weight q4_0 blk.34.attn_k.weight iq2_s +blk.34.attn_output.weight iq3_s blk.34.attn_q.weight iq2_s blk.34.attn_v.weight q4_K -blk.35.ffn_down_exps.weight q4_0 -blk.35.ffn_down_shexp.weight q4_0 +blk.34.ffn_down_exps.weight q4_0 +blk.34.ffn_down_shexp.weight q4_0 blk.35.attn_k.weight iq2_s blk.35.attn_output.weight iq3_s blk.35.attn_q.weight iq2_s blk.35.attn_v.weight q4_K -blk.36.ffn_down_exps.weight q4_0 -blk.36.ffn_down_shexp.weight q4_0 +blk.35.ffn_down_exps.weight q4_0 +blk.35.ffn_down_shexp.weight q4_0 blk.36.attn_k.weight iq2_s blk.36.attn_output.weight iq3_s blk.36.attn_q.weight iq2_s blk.36.attn_v.weight q4_K -blk.37.ffn_down_exps.weight q4_0 -blk.37.ffn_down_shexp.weight q4_0 +blk.36.ffn_down_exps.weight q4_0 +blk.36.ffn_down_shexp.weight q4_0 blk.37.attn_k.weight iq2_s blk.37.attn_output.weight iq3_s blk.37.attn_q.weight iq2_s blk.37.attn_v.weight q4_K -blk.38.ffn_down_exps.weight q4_0 -blk.38.ffn_down_shexp.weight q4_0 +blk.37.ffn_down_exps.weight q4_0 +blk.37.ffn_down_shexp.weight q4_0 blk.38.attn_k.weight iq2_s blk.38.attn_output.weight iq3_s blk.38.attn_q.weight iq2_s blk.38.attn_v.weight q4_K -blk.39.ffn_down_exps.weight q4_0 -blk.39.ffn_down_shexp.weight q4_0 +blk.38.ffn_down_exps.weight q4_0 +blk.38.ffn_down_shexp.weight q4_0 blk.39.attn_k.weight iq2_s blk.39.attn_output.weight iq3_s blk.39.attn_q.weight iq2_s blk.39.attn_v.weight q4_K -blk.40.ffn_down_exps.weight q4_0 -blk.40.ffn_down_shexp.weight q4_0 +blk.39.ffn_down_exps.weight q4_0 +blk.39.ffn_down_shexp.weight q4_0 blk.40.attn_k.weight iq2_s blk.40.attn_output.weight iq3_s blk.40.attn_q.weight iq2_s blk.40.attn_v.weight q4_K -blk.41.ffn_down_exps.weight q4_0 -blk.41.ffn_down_shexp.weight q4_0 +blk.40.ffn_down_exps.weight q4_0 +blk.40.ffn_down_shexp.weight q4_0 blk.41.attn_k.weight iq2_s blk.41.attn_output.weight iq3_s blk.41.attn_q.weight iq2_s blk.41.attn_v.weight q4_K -blk.42.ffn_down_exps.weight q4_0 -blk.42.ffn_down_shexp.weight q4_0 +blk.41.ffn_down_exps.weight q4_0 +blk.41.ffn_down_shexp.weight q4_0 blk.42.attn_k.weight iq2_s +blk.42.attn_output.weight iq3_s blk.42.attn_q.weight iq2_s blk.42.attn_v.weight q4_K -blk.42.attn_output.weight iq3_s -blk.43.ffn_down_exps.weight q4_0 -blk.43.ffn_down_shexp.weight q4_0 +blk.42.ffn_down_exps.weight q4_0 +blk.42.ffn_down_shexp.weight q4_0 blk.43.attn_k.weight iq2_s blk.43.attn_output.weight iq3_s blk.43.attn_q.weight iq2_s blk.43.attn_v.weight q4_K -blk.44.ffn_down_exps.weight q4_0 -blk.44.ffn_down_shexp.weight q4_0 +blk.43.ffn_down_exps.weight q4_0 +blk.43.ffn_down_shexp.weight q4_0 blk.44.attn_k.weight iq2_s blk.44.attn_output.weight iq3_s blk.44.attn_q.weight iq2_s blk.44.attn_v.weight q4_K -output.weight q5_K -blk.45.ffn_down_exps.weight q4_0 -blk.45.ffn_down_shexp.weight q4_0 +blk.44.ffn_down_exps.weight q4_0 +blk.44.ffn_down_shexp.weight q4_0 blk.45.attn_k.weight iq2_s blk.45.attn_output.weight iq3_s blk.45.attn_q.weight iq2_s blk.45.attn_v.weight q4_K +blk.45.ffn_down_exps.weight q4_0 +blk.45.ffn_down_shexp.weight q4_0 [IQ1_S] iq1_s +output.weight q5_K token_embd.weight q2_K -blk.0.ffn_down.weight q4_0 blk.0.attn_output.weight iq2_xxs blk.0.attn_v.weight q4_K -blk.1.ffn_down_exps.weight q4_0 -blk.1.ffn_down_shexp.weight q4_0 +blk.0.ffn_down.weight q4_0 blk.1.attn_output.weight iq2_xxs blk.1.attn_v.weight q4_K -blk.2.ffn_down_exps.weight q4_0 -blk.2.ffn_down_shexp.weight q4_0 +blk.1.ffn_down_exps.weight q4_0 +blk.1.ffn_down_shexp.weight q4_0 blk.2.attn_output.weight iq2_xxs blk.2.attn_v.weight q4_K -blk.3.ffn_down_exps.weight iq4_nl -blk.3.ffn_down_shexp.weight iq4_nl +blk.2.ffn_down_exps.weight q4_0 +blk.2.ffn_down_shexp.weight q4_0 blk.3.attn_output.weight iq2_xxs blk.3.attn_v.weight q4_K -blk.4.ffn_down_exps.weight iq4_nl -blk.4.ffn_down_shexp.weight iq4_nl +blk.3.ffn_down_exps.weight iq4_nl +blk.3.ffn_down_shexp.weight iq4_nl blk.4.attn_output.weight iq2_xxs blk.4.attn_v.weight q4_K -blk.5.ffn_down_exps.weight iq4_nl -blk.5.ffn_down_shexp.weight iq4_nl +blk.4.ffn_down_exps.weight iq4_nl +blk.4.ffn_down_shexp.weight iq4_nl blk.5.attn_output.weight iq2_xxs blk.5.attn_v.weight q4_K -blk.6.ffn_down_exps.weight iq4_nl -blk.6.ffn_down_shexp.weight iq4_nl +blk.5.ffn_down_exps.weight iq4_nl +blk.5.ffn_down_shexp.weight iq4_nl blk.6.attn_output.weight iq2_xxs blk.6.attn_v.weight q4_K -blk.7.ffn_down_exps.weight iq4_nl -blk.7.ffn_down_shexp.weight iq4_nl +blk.6.ffn_down_exps.weight iq4_nl +blk.6.ffn_down_shexp.weight iq4_nl blk.7.attn_output.weight iq2_xxs blk.7.attn_v.weight q4_K -blk.8.ffn_down_exps.weight iq4_nl -blk.8.ffn_down_shexp.weight iq4_nl +blk.7.ffn_down_exps.weight iq4_nl +blk.7.ffn_down_shexp.weight iq4_nl blk.8.attn_output.weight iq2_xxs blk.8.attn_v.weight q4_K -blk.9.ffn_down_exps.weight iq4_nl -blk.9.ffn_down_shexp.weight iq4_nl +blk.8.ffn_down_exps.weight iq4_nl +blk.8.ffn_down_shexp.weight iq4_nl blk.9.attn_output.weight iq2_xxs blk.9.attn_v.weight q4_K -blk.10.ffn_down_exps.weight iq4_nl -blk.10.ffn_down_shexp.weight iq4_nl +blk.9.ffn_down_exps.weight iq4_nl +blk.9.ffn_down_shexp.weight iq4_nl blk.10.attn_output.weight iq2_xxs blk.10.attn_v.weight q4_K -blk.11.ffn_down_exps.weight iq4_nl -blk.11.ffn_down_shexp.weight iq4_nl +blk.10.ffn_down_exps.weight iq4_nl +blk.10.ffn_down_shexp.weight iq4_nl blk.11.attn_output.weight iq2_xxs blk.11.attn_v.weight q4_K -blk.12.ffn_down_exps.weight iq4_nl -blk.12.ffn_down_shexp.weight iq4_nl +blk.11.ffn_down_exps.weight iq4_nl +blk.11.ffn_down_shexp.weight iq4_nl blk.12.attn_output.weight iq2_xxs blk.12.attn_v.weight q4_K -blk.13.ffn_down_exps.weight iq4_nl -blk.13.ffn_down_shexp.weight iq4_nl +blk.12.ffn_down_exps.weight iq4_nl +blk.12.ffn_down_shexp.weight iq4_nl blk.13.attn_output.weight iq2_xxs blk.13.attn_v.weight q4_K -blk.14.ffn_down_exps.weight iq4_nl -blk.14.ffn_down_shexp.weight iq4_nl +blk.13.ffn_down_exps.weight iq4_nl +blk.13.ffn_down_shexp.weight iq4_nl blk.14.attn_output.weight iq2_xxs blk.14.attn_v.weight q4_K -blk.15.ffn_down_exps.weight iq4_nl -blk.15.ffn_down_shexp.weight iq4_nl +blk.14.ffn_down_exps.weight iq4_nl +blk.14.ffn_down_shexp.weight iq4_nl blk.15.attn_output.weight iq2_xxs blk.15.attn_v.weight q4_K -blk.16.ffn_down_exps.weight iq4_nl -blk.16.ffn_down_shexp.weight iq4_nl +blk.15.ffn_down_exps.weight iq4_nl +blk.15.ffn_down_shexp.weight iq4_nl blk.16.attn_output.weight iq2_xxs blk.16.attn_v.weight q4_K -blk.17.ffn_down_exps.weight iq4_nl -blk.17.ffn_down_shexp.weight iq4_nl +blk.16.ffn_down_exps.weight iq4_nl +blk.16.ffn_down_shexp.weight iq4_nl blk.17.attn_output.weight iq2_xxs blk.17.attn_v.weight q4_K -blk.18.ffn_down_exps.weight iq4_nl -blk.18.ffn_down_shexp.weight iq4_nl +blk.17.ffn_down_exps.weight iq4_nl +blk.17.ffn_down_shexp.weight iq4_nl blk.18.attn_output.weight iq2_xxs blk.18.attn_v.weight q4_K -blk.19.ffn_down_exps.weight iq4_nl -blk.19.ffn_down_shexp.weight iq4_nl +blk.18.ffn_down_exps.weight iq4_nl +blk.18.ffn_down_shexp.weight iq4_nl blk.19.attn_output.weight iq2_xxs blk.19.attn_v.weight q4_K -blk.20.ffn_down_exps.weight iq4_nl -blk.20.ffn_down_shexp.weight iq4_nl +blk.19.ffn_down_exps.weight iq4_nl +blk.19.ffn_down_shexp.weight iq4_nl blk.20.attn_output.weight iq2_xxs blk.20.attn_v.weight q4_K -blk.21.ffn_down_exps.weight iq4_nl -blk.21.ffn_down_shexp.weight iq4_nl +blk.20.ffn_down_exps.weight iq4_nl +blk.20.ffn_down_shexp.weight iq4_nl blk.21.attn_output.weight iq2_xxs blk.21.attn_v.weight q4_K -blk.22.ffn_down_exps.weight iq4_nl -blk.22.ffn_down_shexp.weight iq4_nl +blk.21.ffn_down_exps.weight iq4_nl +blk.21.ffn_down_shexp.weight iq4_nl blk.22.attn_output.weight iq2_xxs blk.22.attn_v.weight q4_K -blk.23.ffn_down_exps.weight iq4_nl -blk.23.ffn_down_shexp.weight iq4_nl +blk.22.ffn_down_exps.weight iq4_nl +blk.22.ffn_down_shexp.weight iq4_nl blk.23.attn_output.weight iq2_xxs blk.23.attn_v.weight q4_K -blk.24.ffn_down_exps.weight iq4_nl -blk.24.ffn_down_shexp.weight iq4_nl +blk.23.ffn_down_exps.weight iq4_nl +blk.23.ffn_down_shexp.weight iq4_nl blk.24.attn_output.weight iq2_xxs blk.24.attn_v.weight q4_K -blk.25.ffn_down_exps.weight iq4_nl -blk.25.ffn_down_shexp.weight iq4_nl +blk.24.ffn_down_exps.weight iq4_nl +blk.24.ffn_down_shexp.weight iq4_nl blk.25.attn_output.weight iq2_xxs blk.25.attn_v.weight q4_K -blk.26.ffn_down_exps.weight iq4_nl -blk.26.ffn_down_shexp.weight iq4_nl +blk.25.ffn_down_exps.weight iq4_nl +blk.25.ffn_down_shexp.weight iq4_nl blk.26.attn_output.weight iq2_xxs blk.26.attn_v.weight q4_K -blk.27.ffn_down_exps.weight iq4_nl -blk.27.ffn_down_shexp.weight iq4_nl +blk.26.ffn_down_exps.weight iq4_nl +blk.26.ffn_down_shexp.weight iq4_nl blk.27.attn_output.weight iq2_xxs blk.27.attn_v.weight q4_K -blk.28.ffn_down_exps.weight iq4_nl -blk.28.ffn_down_shexp.weight iq4_nl +blk.27.ffn_down_exps.weight iq4_nl +blk.27.ffn_down_shexp.weight iq4_nl blk.28.attn_output.weight iq2_xxs blk.28.attn_v.weight q4_K -blk.29.ffn_down_exps.weight iq4_nl -blk.29.ffn_down_shexp.weight iq4_nl +blk.28.ffn_down_exps.weight iq4_nl +blk.28.ffn_down_shexp.weight iq4_nl blk.29.attn_output.weight iq2_xxs blk.29.attn_v.weight q4_K -blk.30.ffn_down_exps.weight iq4_nl -blk.30.ffn_down_shexp.weight iq4_nl +blk.29.ffn_down_exps.weight iq4_nl +blk.29.ffn_down_shexp.weight iq4_nl blk.30.attn_output.weight iq2_xxs blk.30.attn_v.weight q4_K -blk.31.ffn_down_exps.weight iq4_nl -blk.31.ffn_down_shexp.weight iq4_nl +blk.30.ffn_down_exps.weight iq4_nl +blk.30.ffn_down_shexp.weight iq4_nl blk.31.attn_output.weight iq2_xxs blk.31.attn_v.weight q4_K -blk.32.ffn_down_exps.weight iq4_nl -blk.32.ffn_down_shexp.weight iq4_nl +blk.31.ffn_down_exps.weight iq4_nl +blk.31.ffn_down_shexp.weight iq4_nl blk.32.attn_output.weight iq2_xxs blk.32.attn_v.weight q4_K -blk.33.ffn_down_exps.weight iq4_nl -blk.33.ffn_down_shexp.weight iq4_nl +blk.32.ffn_down_exps.weight iq4_nl +blk.32.ffn_down_shexp.weight iq4_nl blk.33.attn_output.weight iq2_xxs blk.33.attn_v.weight q4_K -blk.34.ffn_down_exps.weight iq4_nl +blk.33.ffn_down_exps.weight iq4_nl +blk.33.ffn_down_shexp.weight iq4_nl blk.34.attn_output.weight iq2_xxs -blk.34.ffn_down_shexp.weight iq4_nl blk.34.attn_v.weight q4_K -blk.35.ffn_down_exps.weight iq4_nl -blk.35.ffn_down_shexp.weight iq4_nl +blk.34.ffn_down_exps.weight iq4_nl +blk.34.ffn_down_shexp.weight iq4_nl blk.35.attn_output.weight iq2_xxs blk.35.attn_v.weight q4_K -blk.36.ffn_down_exps.weight iq4_nl -blk.36.ffn_down_shexp.weight iq4_nl +blk.35.ffn_down_exps.weight iq4_nl +blk.35.ffn_down_shexp.weight iq4_nl blk.36.attn_output.weight iq2_xxs blk.36.attn_v.weight q4_K -blk.37.ffn_down_exps.weight iq4_nl -blk.37.ffn_down_shexp.weight iq4_nl +blk.36.ffn_down_exps.weight iq4_nl +blk.36.ffn_down_shexp.weight iq4_nl blk.37.attn_output.weight iq2_xxs blk.37.attn_v.weight q4_K -blk.38.ffn_down_exps.weight iq4_nl -blk.38.ffn_down_shexp.weight iq4_nl +blk.37.ffn_down_exps.weight iq4_nl +blk.37.ffn_down_shexp.weight iq4_nl blk.38.attn_output.weight iq2_xxs blk.38.attn_v.weight q4_K -blk.39.ffn_down_exps.weight iq4_nl -blk.39.ffn_down_shexp.weight iq4_nl +blk.38.ffn_down_exps.weight iq4_nl +blk.38.ffn_down_shexp.weight iq4_nl blk.39.attn_output.weight iq2_xxs blk.39.attn_v.weight q4_K -blk.40.ffn_down_exps.weight iq4_nl -blk.40.ffn_down_shexp.weight iq4_nl +blk.39.ffn_down_exps.weight iq4_nl +blk.39.ffn_down_shexp.weight iq4_nl blk.40.attn_output.weight iq2_xxs blk.40.attn_v.weight q4_K -blk.41.ffn_down_exps.weight iq4_nl -blk.41.ffn_down_shexp.weight iq4_nl +blk.40.ffn_down_exps.weight iq4_nl +blk.40.ffn_down_shexp.weight iq4_nl blk.41.attn_output.weight iq2_xxs blk.41.attn_v.weight q4_K +blk.41.ffn_down_exps.weight iq4_nl +blk.41.ffn_down_shexp.weight iq4_nl +blk.42.attn_output.weight iq2_xxs +blk.42.attn_v.weight q4_K blk.42.ffn_down_exps.weight iq4_nl blk.42.ffn_down_shexp.weight iq4_nl -blk.42.attn_v.weight q4_K -blk.42.attn_output.weight iq2_xxs -blk.43.ffn_down_exps.weight iq4_nl -blk.43.ffn_down_shexp.weight iq4_nl blk.43.attn_output.weight iq2_xxs blk.43.attn_v.weight q4_K -blk.44.ffn_down_exps.weight iq4_nl -blk.44.ffn_down_shexp.weight iq4_nl +blk.43.ffn_down_exps.weight iq4_nl +blk.43.ffn_down_shexp.weight iq4_nl blk.44.attn_output.weight iq2_xxs blk.44.attn_v.weight q4_K -output.weight q5_K -blk.45.ffn_down_exps.weight iq4_nl -blk.45.ffn_down_shexp.weight iq4_nl +blk.44.ffn_down_exps.weight iq4_nl +blk.44.ffn_down_shexp.weight iq4_nl blk.45.attn_output.weight iq2_xxs blk.45.attn_v.weight q4_K +blk.45.ffn_down_exps.weight iq4_nl +blk.45.ffn_down_shexp.weight iq4_nl [IQ4_NL] iq4_nl -blk.0.ffn_down.weight q5_1 +output.weight q6_K blk.0.attn_v.weight q5_K +blk.0.ffn_down.weight q5_1 +blk.1.attn_v.weight q5_K blk.1.ffn_down_exps.weight q5_1 blk.1.ffn_down_shexp.weight q5_1 -blk.1.attn_v.weight q5_K +blk.2.attn_v.weight q5_K blk.2.ffn_down_exps.weight q5_1 blk.2.ffn_down_shexp.weight q5_1 -blk.2.attn_v.weight q5_K +blk.3.attn_v.weight q5_K blk.3.ffn_down_exps.weight q5_1 blk.3.ffn_down_shexp.weight q5_1 -blk.3.attn_v.weight q5_K +blk.4.attn_v.weight q5_K blk.4.ffn_down_exps.weight q5_1 blk.4.ffn_down_shexp.weight q5_1 -blk.4.attn_v.weight q5_K blk.5.attn_v.weight q5_K blk.6.attn_v.weight q5_K blk.7.attn_v.weight q5_K @@ -2503,1039 +2504,1039 @@ blk.41.attn_v.weight q5_K blk.42.attn_v.weight q5_K blk.43.attn_v.weight q5_K blk.44.attn_v.weight q5_K -output.weight q6_K blk.45.attn_v.weight q5_K [IQ3_S] iq3_s -blk.0.ffn_down.weight iq4_nl +output.weight q6_K blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight iq4_nl +blk.1.attn_v.weight q4_K blk.1.ffn_down_exps.weight iq4_nl blk.1.ffn_down_shexp.weight iq4_nl -blk.1.attn_v.weight q4_K +blk.2.attn_v.weight q4_K blk.2.ffn_down_exps.weight iq4_nl blk.2.ffn_down_shexp.weight iq4_nl -blk.2.attn_v.weight q4_K +blk.3.attn_v.weight q4_K blk.3.ffn_down_exps.weight iq4_nl blk.3.ffn_down_shexp.weight iq4_nl -blk.3.attn_v.weight q4_K +blk.4.attn_v.weight q4_K blk.4.ffn_down_exps.weight iq4_nl blk.4.ffn_down_shexp.weight iq4_nl -blk.4.attn_v.weight q4_K +blk.5.attn_v.weight q4_K blk.5.ffn_down_exps.weight iq4_nl blk.5.ffn_down_shexp.weight iq4_nl -blk.5.attn_v.weight q4_K +blk.6.attn_v.weight q4_K blk.6.ffn_down_exps.weight iq4_nl blk.6.ffn_down_shexp.weight iq4_nl -blk.6.attn_v.weight q4_K +blk.7.attn_v.weight q4_K blk.7.ffn_down_exps.weight iq4_nl blk.7.ffn_down_shexp.weight iq4_nl -blk.7.attn_v.weight q4_K +blk.8.attn_v.weight q4_K blk.8.ffn_down_exps.weight iq4_nl blk.8.ffn_down_shexp.weight iq4_nl -blk.8.attn_v.weight q4_K +blk.9.attn_v.weight q4_K blk.9.ffn_down_exps.weight iq4_nl blk.9.ffn_down_shexp.weight iq4_nl -blk.9.attn_v.weight q4_K +blk.10.attn_v.weight q4_K blk.10.ffn_down_exps.weight iq4_nl blk.10.ffn_down_shexp.weight iq4_nl -blk.10.attn_v.weight q4_K +blk.11.attn_v.weight q4_K blk.11.ffn_down_exps.weight iq4_nl blk.11.ffn_down_shexp.weight iq4_nl -blk.11.attn_v.weight q4_K +blk.12.attn_v.weight q4_K blk.12.ffn_down_exps.weight iq4_nl blk.12.ffn_down_shexp.weight iq4_nl -blk.12.attn_v.weight q4_K +blk.13.attn_v.weight q4_K blk.13.ffn_down_exps.weight iq4_nl blk.13.ffn_down_shexp.weight iq4_nl -blk.13.attn_v.weight q4_K +blk.14.attn_v.weight q4_K blk.14.ffn_down_exps.weight iq4_nl blk.14.ffn_down_shexp.weight iq4_nl -blk.14.attn_v.weight q4_K +blk.15.attn_v.weight q4_K blk.15.ffn_down_exps.weight iq4_nl blk.15.ffn_down_shexp.weight iq4_nl -blk.15.attn_v.weight q4_K +blk.16.attn_v.weight q4_K blk.16.ffn_down_exps.weight iq4_nl blk.16.ffn_down_shexp.weight iq4_nl -blk.16.attn_v.weight q4_K +blk.17.attn_v.weight q4_K blk.17.ffn_down_exps.weight iq4_nl blk.17.ffn_down_shexp.weight iq4_nl -blk.17.attn_v.weight q4_K +blk.18.attn_v.weight q4_K blk.18.ffn_down_exps.weight iq4_nl blk.18.ffn_down_shexp.weight iq4_nl -blk.18.attn_v.weight q4_K +blk.19.attn_v.weight q4_K blk.19.ffn_down_exps.weight iq4_nl blk.19.ffn_down_shexp.weight iq4_nl -blk.19.attn_v.weight q4_K +blk.20.attn_v.weight q4_K blk.20.ffn_down_exps.weight iq4_nl blk.20.ffn_down_shexp.weight iq4_nl -blk.20.attn_v.weight q4_K +blk.21.attn_v.weight q4_K blk.21.ffn_down_exps.weight iq4_nl blk.21.ffn_down_shexp.weight iq4_nl -blk.21.attn_v.weight q4_K +blk.22.attn_v.weight q4_K blk.22.ffn_down_exps.weight iq4_nl blk.22.ffn_down_shexp.weight iq4_nl -blk.22.attn_v.weight q4_K +blk.23.attn_v.weight q4_K blk.23.ffn_down_exps.weight iq4_nl blk.23.ffn_down_shexp.weight iq4_nl -blk.23.attn_v.weight q4_K +blk.24.attn_v.weight q4_K blk.24.ffn_down_exps.weight iq4_nl blk.24.ffn_down_shexp.weight iq4_nl -blk.24.attn_v.weight q4_K +blk.25.attn_v.weight q4_K blk.25.ffn_down_exps.weight iq4_nl blk.25.ffn_down_shexp.weight iq4_nl -blk.25.attn_v.weight q4_K +blk.26.attn_v.weight q4_K blk.26.ffn_down_exps.weight iq4_nl blk.26.ffn_down_shexp.weight iq4_nl -blk.26.attn_v.weight q4_K +blk.27.attn_v.weight q4_K blk.27.ffn_down_exps.weight iq4_nl blk.27.ffn_down_shexp.weight iq4_nl -blk.27.attn_v.weight q4_K +blk.28.attn_v.weight q4_K blk.28.ffn_down_exps.weight iq4_nl blk.28.ffn_down_shexp.weight iq4_nl -blk.28.attn_v.weight q4_K +blk.29.attn_v.weight q4_K blk.29.ffn_down_exps.weight iq4_nl blk.29.ffn_down_shexp.weight iq4_nl -blk.29.attn_v.weight q4_K +blk.30.attn_v.weight q4_K blk.30.ffn_down_exps.weight iq4_nl blk.30.ffn_down_shexp.weight iq4_nl -blk.30.attn_v.weight q4_K +blk.31.attn_v.weight q4_K blk.31.ffn_down_exps.weight iq4_nl blk.31.ffn_down_shexp.weight iq4_nl -blk.31.attn_v.weight q4_K +blk.32.attn_v.weight q4_K blk.32.ffn_down_exps.weight iq4_nl blk.32.ffn_down_shexp.weight iq4_nl -blk.32.attn_v.weight q4_K +blk.33.attn_v.weight q4_K blk.33.ffn_down_exps.weight iq4_nl blk.33.ffn_down_shexp.weight iq4_nl -blk.33.attn_v.weight q4_K +blk.34.attn_v.weight q4_K blk.34.ffn_down_exps.weight iq4_nl blk.34.ffn_down_shexp.weight iq4_nl -blk.34.attn_v.weight q4_K +blk.35.attn_v.weight q4_K blk.35.ffn_down_exps.weight iq4_nl blk.35.ffn_down_shexp.weight iq4_nl -blk.35.attn_v.weight q4_K +blk.36.attn_v.weight q4_K blk.36.ffn_down_exps.weight iq4_nl blk.36.ffn_down_shexp.weight iq4_nl -blk.36.attn_v.weight q4_K +blk.37.attn_v.weight q4_K blk.37.ffn_down_exps.weight iq4_nl blk.37.ffn_down_shexp.weight iq4_nl -blk.37.attn_v.weight q4_K +blk.38.attn_v.weight q4_K blk.38.ffn_down_exps.weight iq4_nl blk.38.ffn_down_shexp.weight iq4_nl -blk.38.attn_v.weight q4_K +blk.39.attn_v.weight q4_K blk.39.ffn_down_exps.weight iq4_nl blk.39.ffn_down_shexp.weight iq4_nl -blk.39.attn_v.weight q4_K +blk.40.attn_v.weight q4_K blk.40.ffn_down_exps.weight iq4_nl blk.40.ffn_down_shexp.weight iq4_nl -blk.40.attn_v.weight q4_K +blk.41.attn_v.weight q4_K blk.41.ffn_down_exps.weight iq4_nl blk.41.ffn_down_shexp.weight iq4_nl -blk.41.attn_v.weight q4_K +blk.42.attn_v.weight q4_K blk.42.ffn_down_exps.weight iq4_nl blk.42.ffn_down_shexp.weight iq4_nl -blk.42.attn_v.weight q4_K +blk.43.attn_v.weight q4_K blk.43.ffn_down_exps.weight iq4_nl blk.43.ffn_down_shexp.weight iq4_nl -blk.43.attn_v.weight q4_K +blk.44.attn_v.weight q4_K blk.44.ffn_down_exps.weight iq4_nl blk.44.ffn_down_shexp.weight iq4_nl -blk.44.attn_v.weight q4_K -output.weight q6_K +blk.45.attn_v.weight q4_K blk.45.ffn_down_exps.weight iq4_nl blk.45.ffn_down_shexp.weight iq4_nl -blk.45.attn_v.weight q4_K [IQ3_M] iq3_s -blk.0.ffn_down.weight q5_0 +output.weight q6_K blk.0.attn_output.weight q4_K blk.0.attn_v.weight q4_K -blk.1.ffn_down_exps.weight q5_0 -blk.1.ffn_down_shexp.weight q5_0 +blk.0.ffn_down.weight q5_0 blk.1.attn_output.weight q4_K blk.1.attn_v.weight q4_K -blk.2.ffn_down_exps.weight q5_0 -blk.2.ffn_down_shexp.weight q5_0 +blk.1.ffn_down_exps.weight q5_0 +blk.1.ffn_down_shexp.weight q5_0 blk.2.attn_output.weight q4_K blk.2.attn_v.weight q4_K -blk.3.ffn_down_exps.weight q5_0 -blk.3.ffn_down_shexp.weight q5_0 +blk.2.ffn_down_exps.weight q5_0 +blk.2.ffn_down_shexp.weight q5_0 blk.3.attn_output.weight q4_K blk.3.attn_v.weight q4_K -blk.4.ffn_down_exps.weight q5_0 -blk.4.ffn_down_shexp.weight q5_0 +blk.3.ffn_down_exps.weight q5_0 +blk.3.ffn_down_shexp.weight q5_0 blk.4.attn_output.weight q4_K blk.4.attn_v.weight q4_K -blk.5.ffn_down_exps.weight iq4_nl -blk.5.ffn_down_shexp.weight iq4_nl +blk.4.ffn_down_exps.weight q5_0 +blk.4.ffn_down_shexp.weight q5_0 blk.5.attn_output.weight q4_K blk.5.attn_v.weight q4_K -blk.6.ffn_down_exps.weight iq4_nl -blk.6.ffn_down_shexp.weight iq4_nl +blk.5.ffn_down_exps.weight iq4_nl +blk.5.ffn_down_shexp.weight iq4_nl blk.6.attn_output.weight q4_K blk.6.attn_v.weight q4_K -blk.7.ffn_down_exps.weight iq4_nl -blk.7.ffn_down_shexp.weight iq4_nl +blk.6.ffn_down_exps.weight iq4_nl +blk.6.ffn_down_shexp.weight iq4_nl blk.7.attn_output.weight q4_K blk.7.attn_v.weight q4_K -blk.8.ffn_down_exps.weight iq4_nl -blk.8.ffn_down_shexp.weight iq4_nl +blk.7.ffn_down_exps.weight iq4_nl +blk.7.ffn_down_shexp.weight iq4_nl blk.8.attn_output.weight q4_K blk.8.attn_v.weight q4_K -blk.9.ffn_down_exps.weight iq4_nl -blk.9.ffn_down_shexp.weight iq4_nl +blk.8.ffn_down_exps.weight iq4_nl +blk.8.ffn_down_shexp.weight iq4_nl blk.9.attn_output.weight q4_K blk.9.attn_v.weight q4_K -blk.10.ffn_down_exps.weight iq4_nl -blk.10.ffn_down_shexp.weight iq4_nl +blk.9.ffn_down_exps.weight iq4_nl +blk.9.ffn_down_shexp.weight iq4_nl blk.10.attn_output.weight q4_K blk.10.attn_v.weight q4_K -blk.11.ffn_down_exps.weight iq4_nl -blk.11.ffn_down_shexp.weight iq4_nl +blk.10.ffn_down_exps.weight iq4_nl +blk.10.ffn_down_shexp.weight iq4_nl blk.11.attn_output.weight q4_K blk.11.attn_v.weight q4_K -blk.12.ffn_down_exps.weight iq4_nl -blk.12.ffn_down_shexp.weight iq4_nl +blk.11.ffn_down_exps.weight iq4_nl +blk.11.ffn_down_shexp.weight iq4_nl blk.12.attn_output.weight q4_K blk.12.attn_v.weight q4_K -blk.13.ffn_down_exps.weight iq4_nl -blk.13.ffn_down_shexp.weight iq4_nl +blk.12.ffn_down_exps.weight iq4_nl +blk.12.ffn_down_shexp.weight iq4_nl blk.13.attn_output.weight q4_K blk.13.attn_v.weight q4_K -blk.14.ffn_down_exps.weight iq4_nl -blk.14.ffn_down_shexp.weight iq4_nl +blk.13.ffn_down_exps.weight iq4_nl +blk.13.ffn_down_shexp.weight iq4_nl blk.14.attn_output.weight q4_K blk.14.attn_v.weight q4_K -blk.15.ffn_down_exps.weight iq4_nl -blk.15.ffn_down_shexp.weight iq4_nl +blk.14.ffn_down_exps.weight iq4_nl +blk.14.ffn_down_shexp.weight iq4_nl blk.15.attn_output.weight q4_K blk.15.attn_v.weight q4_K -blk.16.ffn_down_exps.weight iq4_nl -blk.16.ffn_down_shexp.weight iq4_nl +blk.15.ffn_down_exps.weight iq4_nl +blk.15.ffn_down_shexp.weight iq4_nl blk.16.attn_output.weight q4_K blk.16.attn_v.weight q4_K -blk.17.ffn_down_exps.weight iq4_nl -blk.17.ffn_down_shexp.weight iq4_nl +blk.16.ffn_down_exps.weight iq4_nl +blk.16.ffn_down_shexp.weight iq4_nl blk.17.attn_output.weight q4_K blk.17.attn_v.weight q4_K -blk.18.ffn_down_exps.weight iq4_nl -blk.18.ffn_down_shexp.weight iq4_nl +blk.17.ffn_down_exps.weight iq4_nl +blk.17.ffn_down_shexp.weight iq4_nl blk.18.attn_output.weight q4_K blk.18.attn_v.weight q4_K -blk.19.ffn_down_exps.weight iq4_nl -blk.19.ffn_down_shexp.weight iq4_nl +blk.18.ffn_down_exps.weight iq4_nl +blk.18.ffn_down_shexp.weight iq4_nl blk.19.attn_output.weight q4_K blk.19.attn_v.weight q4_K -blk.20.ffn_down_exps.weight iq4_nl -blk.20.ffn_down_shexp.weight iq4_nl +blk.19.ffn_down_exps.weight iq4_nl +blk.19.ffn_down_shexp.weight iq4_nl blk.20.attn_output.weight q4_K blk.20.attn_v.weight q4_K -blk.21.ffn_down_exps.weight iq4_nl -blk.21.ffn_down_shexp.weight iq4_nl +blk.20.ffn_down_exps.weight iq4_nl +blk.20.ffn_down_shexp.weight iq4_nl blk.21.attn_output.weight q4_K blk.21.attn_v.weight q4_K -blk.22.ffn_down_exps.weight iq4_nl -blk.22.ffn_down_shexp.weight iq4_nl +blk.21.ffn_down_exps.weight iq4_nl +blk.21.ffn_down_shexp.weight iq4_nl blk.22.attn_output.weight q4_K blk.22.attn_v.weight q4_K -blk.23.ffn_down_exps.weight iq4_nl -blk.23.ffn_down_shexp.weight iq4_nl +blk.22.ffn_down_exps.weight iq4_nl +blk.22.ffn_down_shexp.weight iq4_nl blk.23.attn_output.weight q4_K blk.23.attn_v.weight q4_K -blk.24.ffn_down_exps.weight iq4_nl -blk.24.ffn_down_shexp.weight iq4_nl +blk.23.ffn_down_exps.weight iq4_nl +blk.23.ffn_down_shexp.weight iq4_nl blk.24.attn_output.weight q4_K blk.24.attn_v.weight q4_K -blk.25.ffn_down_exps.weight iq4_nl -blk.25.ffn_down_shexp.weight iq4_nl +blk.24.ffn_down_exps.weight iq4_nl +blk.24.ffn_down_shexp.weight iq4_nl blk.25.attn_output.weight q4_K blk.25.attn_v.weight q4_K -blk.26.ffn_down_exps.weight iq4_nl -blk.26.ffn_down_shexp.weight iq4_nl +blk.25.ffn_down_exps.weight iq4_nl +blk.25.ffn_down_shexp.weight iq4_nl blk.26.attn_output.weight q4_K blk.26.attn_v.weight q4_K -blk.27.ffn_down_exps.weight iq4_nl -blk.27.ffn_down_shexp.weight iq4_nl +blk.26.ffn_down_exps.weight iq4_nl +blk.26.ffn_down_shexp.weight iq4_nl blk.27.attn_output.weight q4_K blk.27.attn_v.weight q4_K -blk.28.ffn_down_exps.weight iq4_nl -blk.28.ffn_down_shexp.weight iq4_nl +blk.27.ffn_down_exps.weight iq4_nl +blk.27.ffn_down_shexp.weight iq4_nl blk.28.attn_output.weight q4_K blk.28.attn_v.weight q4_K -blk.29.ffn_down_exps.weight iq4_nl -blk.29.ffn_down_shexp.weight iq4_nl +blk.28.ffn_down_exps.weight iq4_nl +blk.28.ffn_down_shexp.weight iq4_nl blk.29.attn_output.weight q4_K blk.29.attn_v.weight q4_K -blk.30.ffn_down_exps.weight iq4_nl -blk.30.ffn_down_shexp.weight iq4_nl +blk.29.ffn_down_exps.weight iq4_nl +blk.29.ffn_down_shexp.weight iq4_nl blk.30.attn_output.weight q4_K blk.30.attn_v.weight q4_K -blk.31.ffn_down_exps.weight iq4_nl -blk.31.ffn_down_shexp.weight iq4_nl +blk.30.ffn_down_exps.weight iq4_nl +blk.30.ffn_down_shexp.weight iq4_nl blk.31.attn_output.weight q4_K blk.31.attn_v.weight q4_K -blk.32.ffn_down_exps.weight iq4_nl -blk.32.ffn_down_shexp.weight iq4_nl +blk.31.ffn_down_exps.weight iq4_nl +blk.31.ffn_down_shexp.weight iq4_nl blk.32.attn_output.weight q4_K blk.32.attn_v.weight q4_K -blk.33.ffn_down_exps.weight iq4_nl -blk.33.ffn_down_shexp.weight iq4_nl +blk.32.ffn_down_exps.weight iq4_nl +blk.32.ffn_down_shexp.weight iq4_nl blk.33.attn_output.weight q4_K blk.33.attn_v.weight q4_K -blk.34.ffn_down_exps.weight iq4_nl +blk.33.ffn_down_exps.weight iq4_nl +blk.33.ffn_down_shexp.weight iq4_nl blk.34.attn_output.weight q4_K -blk.34.ffn_down_shexp.weight iq4_nl blk.34.attn_v.weight q4_K -blk.35.ffn_down_exps.weight iq4_nl -blk.35.ffn_down_shexp.weight iq4_nl +blk.34.ffn_down_exps.weight iq4_nl +blk.34.ffn_down_shexp.weight iq4_nl blk.35.attn_output.weight q4_K blk.35.attn_v.weight q4_K -blk.36.ffn_down_exps.weight iq4_nl -blk.36.ffn_down_shexp.weight iq4_nl +blk.35.ffn_down_exps.weight iq4_nl +blk.35.ffn_down_shexp.weight iq4_nl blk.36.attn_output.weight q4_K blk.36.attn_v.weight q4_K -blk.37.ffn_down_exps.weight iq4_nl -blk.37.ffn_down_shexp.weight iq4_nl +blk.36.ffn_down_exps.weight iq4_nl +blk.36.ffn_down_shexp.weight iq4_nl blk.37.attn_output.weight q4_K blk.37.attn_v.weight q4_K -blk.38.ffn_down_exps.weight iq4_nl -blk.38.ffn_down_shexp.weight iq4_nl +blk.37.ffn_down_exps.weight iq4_nl +blk.37.ffn_down_shexp.weight iq4_nl blk.38.attn_output.weight q4_K blk.38.attn_v.weight q4_K -blk.39.ffn_down_exps.weight iq4_nl -blk.39.ffn_down_shexp.weight iq4_nl +blk.38.ffn_down_exps.weight iq4_nl +blk.38.ffn_down_shexp.weight iq4_nl blk.39.attn_output.weight q4_K blk.39.attn_v.weight q4_K -blk.40.ffn_down_exps.weight iq4_nl -blk.40.ffn_down_shexp.weight iq4_nl +blk.39.ffn_down_exps.weight iq4_nl +blk.39.ffn_down_shexp.weight iq4_nl blk.40.attn_output.weight q4_K blk.40.attn_v.weight q4_K -blk.41.ffn_down_exps.weight iq4_nl -blk.41.ffn_down_shexp.weight iq4_nl +blk.40.ffn_down_exps.weight iq4_nl +blk.40.ffn_down_shexp.weight iq4_nl blk.41.attn_output.weight q4_K blk.41.attn_v.weight q4_K +blk.41.ffn_down_exps.weight iq4_nl +blk.41.ffn_down_shexp.weight iq4_nl +blk.42.attn_output.weight q4_K +blk.42.attn_v.weight q4_K blk.42.ffn_down_exps.weight iq4_nl blk.42.ffn_down_shexp.weight iq4_nl -blk.42.attn_v.weight q4_K -blk.42.attn_output.weight q4_K -blk.43.ffn_down_exps.weight iq4_nl -blk.43.ffn_down_shexp.weight iq4_nl blk.43.attn_output.weight q4_K blk.43.attn_v.weight q4_K -blk.44.ffn_down_exps.weight iq4_nl -blk.44.ffn_down_shexp.weight iq4_nl +blk.43.ffn_down_exps.weight iq4_nl +blk.43.ffn_down_shexp.weight iq4_nl blk.44.attn_output.weight q4_K blk.44.attn_v.weight q4_K -output.weight q6_K -blk.45.ffn_down_exps.weight iq4_nl -blk.45.ffn_down_shexp.weight iq4_nl +blk.44.ffn_down_exps.weight iq4_nl +blk.44.ffn_down_shexp.weight iq4_nl blk.45.attn_output.weight q4_K blk.45.attn_v.weight q4_K +blk.45.ffn_down_exps.weight iq4_nl +blk.45.ffn_down_shexp.weight iq4_nl [IQ2_S] iq2_xs +output.weight q5_K token_embd.weight iq3_s -blk.0.ffn_down.weight iq4_nl blk.0.attn_output.weight iq3_s blk.0.attn_v.weight q4_K -blk.1.ffn_down_exps.weight iq4_nl -blk.1.ffn_down_shexp.weight iq4_nl +blk.0.ffn_down.weight iq4_nl blk.1.attn_output.weight iq3_s blk.1.attn_v.weight q4_K -blk.2.ffn_down_exps.weight iq4_nl -blk.2.ffn_down_shexp.weight iq4_nl +blk.1.ffn_down_exps.weight iq4_nl +blk.1.ffn_down_shexp.weight iq4_nl blk.2.attn_output.weight iq3_s blk.2.attn_v.weight q4_K -blk.3.ffn_down_exps.weight iq4_nl -blk.3.ffn_down_shexp.weight iq4_nl +blk.2.ffn_down_exps.weight iq4_nl +blk.2.ffn_down_shexp.weight iq4_nl blk.3.attn_output.weight iq3_s blk.3.attn_v.weight q4_K -blk.4.ffn_down_exps.weight iq4_nl -blk.4.ffn_down_shexp.weight iq4_nl +blk.3.ffn_down_exps.weight iq4_nl +blk.3.ffn_down_shexp.weight iq4_nl blk.4.attn_output.weight iq3_s blk.4.attn_v.weight q4_K -blk.5.ffn_down_exps.weight iq4_nl -blk.5.ffn_down_shexp.weight iq4_nl +blk.4.ffn_down_exps.weight iq4_nl +blk.4.ffn_down_shexp.weight iq4_nl blk.5.attn_output.weight iq3_s blk.5.attn_v.weight q4_K -blk.6.ffn_down_exps.weight iq4_nl -blk.6.ffn_down_shexp.weight iq4_nl +blk.5.ffn_down_exps.weight iq4_nl +blk.5.ffn_down_shexp.weight iq4_nl blk.6.attn_output.weight iq3_s blk.6.attn_v.weight q4_K -blk.7.ffn_down_exps.weight iq4_nl -blk.7.ffn_down_shexp.weight iq4_nl +blk.6.ffn_down_exps.weight iq4_nl +blk.6.ffn_down_shexp.weight iq4_nl blk.7.attn_output.weight iq3_s blk.7.attn_v.weight q4_K -blk.8.ffn_down_exps.weight iq4_nl -blk.8.ffn_down_shexp.weight iq4_nl +blk.7.ffn_down_exps.weight iq4_nl +blk.7.ffn_down_shexp.weight iq4_nl blk.8.attn_output.weight iq3_s blk.8.attn_v.weight q4_K -blk.9.ffn_down_exps.weight iq4_nl -blk.9.ffn_down_shexp.weight iq4_nl +blk.8.ffn_down_exps.weight iq4_nl +blk.8.ffn_down_shexp.weight iq4_nl blk.9.attn_output.weight iq3_s blk.9.attn_v.weight q4_K -blk.10.ffn_down_exps.weight iq4_nl -blk.10.ffn_down_shexp.weight iq4_nl +blk.9.ffn_down_exps.weight iq4_nl +blk.9.ffn_down_shexp.weight iq4_nl blk.10.attn_output.weight iq3_s blk.10.attn_v.weight q4_K -blk.11.ffn_down_exps.weight iq4_nl -blk.11.ffn_down_shexp.weight iq4_nl +blk.10.ffn_down_exps.weight iq4_nl +blk.10.ffn_down_shexp.weight iq4_nl blk.11.attn_output.weight iq3_s blk.11.attn_v.weight q4_K -blk.12.ffn_down_exps.weight iq4_nl -blk.12.ffn_down_shexp.weight iq4_nl +blk.11.ffn_down_exps.weight iq4_nl +blk.11.ffn_down_shexp.weight iq4_nl blk.12.attn_output.weight iq3_s blk.12.attn_v.weight q4_K -blk.13.ffn_down_exps.weight iq4_nl -blk.13.ffn_down_shexp.weight iq4_nl +blk.12.ffn_down_exps.weight iq4_nl +blk.12.ffn_down_shexp.weight iq4_nl blk.13.attn_output.weight iq3_s blk.13.attn_v.weight q4_K -blk.14.ffn_down_exps.weight iq4_nl -blk.14.ffn_down_shexp.weight iq4_nl +blk.13.ffn_down_exps.weight iq4_nl +blk.13.ffn_down_shexp.weight iq4_nl blk.14.attn_output.weight iq3_s blk.14.attn_v.weight q4_K -blk.15.ffn_down_exps.weight iq4_nl -blk.15.ffn_down_shexp.weight iq4_nl +blk.14.ffn_down_exps.weight iq4_nl +blk.14.ffn_down_shexp.weight iq4_nl blk.15.attn_output.weight iq3_s blk.15.attn_v.weight q4_K -blk.16.ffn_down_exps.weight iq4_nl -blk.16.ffn_down_shexp.weight iq4_nl +blk.15.ffn_down_exps.weight iq4_nl +blk.15.ffn_down_shexp.weight iq4_nl blk.16.attn_output.weight iq3_s blk.16.attn_v.weight q4_K -blk.17.ffn_down_exps.weight iq4_nl -blk.17.ffn_down_shexp.weight iq4_nl +blk.16.ffn_down_exps.weight iq4_nl +blk.16.ffn_down_shexp.weight iq4_nl blk.17.attn_output.weight iq3_s blk.17.attn_v.weight q4_K -blk.18.ffn_down_exps.weight iq4_nl -blk.18.ffn_down_shexp.weight iq4_nl +blk.17.ffn_down_exps.weight iq4_nl +blk.17.ffn_down_shexp.weight iq4_nl blk.18.attn_output.weight iq3_s blk.18.attn_v.weight q4_K -blk.19.ffn_down_exps.weight iq4_nl -blk.19.ffn_down_shexp.weight iq4_nl +blk.18.ffn_down_exps.weight iq4_nl +blk.18.ffn_down_shexp.weight iq4_nl blk.19.attn_output.weight iq3_s blk.19.attn_v.weight q4_K -blk.20.ffn_down_exps.weight iq4_nl -blk.20.ffn_down_shexp.weight iq4_nl +blk.19.ffn_down_exps.weight iq4_nl +blk.19.ffn_down_shexp.weight iq4_nl blk.20.attn_output.weight iq3_s blk.20.attn_v.weight q4_K -blk.21.ffn_down_exps.weight iq4_nl -blk.21.ffn_down_shexp.weight iq4_nl +blk.20.ffn_down_exps.weight iq4_nl +blk.20.ffn_down_shexp.weight iq4_nl blk.21.attn_output.weight iq3_s blk.21.attn_v.weight q4_K -blk.22.ffn_down_exps.weight iq4_nl -blk.22.ffn_down_shexp.weight iq4_nl +blk.21.ffn_down_exps.weight iq4_nl +blk.21.ffn_down_shexp.weight iq4_nl blk.22.attn_output.weight iq3_s blk.22.attn_v.weight q4_K -blk.23.ffn_down_exps.weight iq4_nl -blk.23.ffn_down_shexp.weight iq4_nl +blk.22.ffn_down_exps.weight iq4_nl +blk.22.ffn_down_shexp.weight iq4_nl blk.23.attn_output.weight iq3_s blk.23.attn_v.weight q4_K -blk.24.ffn_down_exps.weight iq4_nl -blk.24.ffn_down_shexp.weight iq4_nl +blk.23.ffn_down_exps.weight iq4_nl +blk.23.ffn_down_shexp.weight iq4_nl blk.24.attn_output.weight iq3_s blk.24.attn_v.weight q4_K -blk.25.ffn_down_exps.weight iq4_nl -blk.25.ffn_down_shexp.weight iq4_nl +blk.24.ffn_down_exps.weight iq4_nl +blk.24.ffn_down_shexp.weight iq4_nl blk.25.attn_output.weight iq3_s blk.25.attn_v.weight q4_K -blk.26.ffn_down_exps.weight iq4_nl -blk.26.ffn_down_shexp.weight iq4_nl +blk.25.ffn_down_exps.weight iq4_nl +blk.25.ffn_down_shexp.weight iq4_nl blk.26.attn_output.weight iq3_s blk.26.attn_v.weight q4_K -blk.27.ffn_down_exps.weight iq4_nl -blk.27.ffn_down_shexp.weight iq4_nl +blk.26.ffn_down_exps.weight iq4_nl +blk.26.ffn_down_shexp.weight iq4_nl blk.27.attn_output.weight iq3_s blk.27.attn_v.weight q4_K -blk.28.ffn_down_exps.weight iq4_nl -blk.28.ffn_down_shexp.weight iq4_nl +blk.27.ffn_down_exps.weight iq4_nl +blk.27.ffn_down_shexp.weight iq4_nl blk.28.attn_output.weight iq3_s blk.28.attn_v.weight q4_K -blk.29.ffn_down_exps.weight iq4_nl -blk.29.ffn_down_shexp.weight iq4_nl +blk.28.ffn_down_exps.weight iq4_nl +blk.28.ffn_down_shexp.weight iq4_nl blk.29.attn_output.weight iq3_s blk.29.attn_v.weight q4_K -blk.30.ffn_down_exps.weight iq4_nl -blk.30.ffn_down_shexp.weight iq4_nl +blk.29.ffn_down_exps.weight iq4_nl +blk.29.ffn_down_shexp.weight iq4_nl blk.30.attn_output.weight iq3_s blk.30.attn_v.weight q4_K -blk.31.ffn_down_exps.weight iq4_nl -blk.31.ffn_down_shexp.weight iq4_nl +blk.30.ffn_down_exps.weight iq4_nl +blk.30.ffn_down_shexp.weight iq4_nl blk.31.attn_output.weight iq3_s blk.31.attn_v.weight q4_K -blk.32.ffn_down_exps.weight iq4_nl -blk.32.ffn_down_shexp.weight iq4_nl +blk.31.ffn_down_exps.weight iq4_nl +blk.31.ffn_down_shexp.weight iq4_nl blk.32.attn_output.weight iq3_s blk.32.attn_v.weight q4_K -blk.33.ffn_down_exps.weight iq4_nl -blk.33.ffn_down_shexp.weight iq4_nl +blk.32.ffn_down_exps.weight iq4_nl +blk.32.ffn_down_shexp.weight iq4_nl blk.33.attn_output.weight iq3_s blk.33.attn_v.weight q4_K -blk.34.ffn_down_exps.weight iq4_nl +blk.33.ffn_down_exps.weight iq4_nl +blk.33.ffn_down_shexp.weight iq4_nl blk.34.attn_output.weight iq3_s -blk.34.ffn_down_shexp.weight iq4_nl blk.34.attn_v.weight q4_K -blk.35.ffn_down_exps.weight iq4_nl -blk.35.ffn_down_shexp.weight iq4_nl +blk.34.ffn_down_exps.weight iq4_nl +blk.34.ffn_down_shexp.weight iq4_nl blk.35.attn_output.weight iq3_s blk.35.attn_v.weight q4_K -blk.36.ffn_down_exps.weight iq4_nl -blk.36.ffn_down_shexp.weight iq4_nl +blk.35.ffn_down_exps.weight iq4_nl +blk.35.ffn_down_shexp.weight iq4_nl blk.36.attn_output.weight iq3_s blk.36.attn_v.weight q4_K -blk.37.ffn_down_exps.weight iq4_nl -blk.37.ffn_down_shexp.weight iq4_nl +blk.36.ffn_down_exps.weight iq4_nl +blk.36.ffn_down_shexp.weight iq4_nl blk.37.attn_output.weight iq3_s blk.37.attn_v.weight q4_K -blk.38.ffn_down_exps.weight iq4_nl -blk.38.ffn_down_shexp.weight iq4_nl +blk.37.ffn_down_exps.weight iq4_nl +blk.37.ffn_down_shexp.weight iq4_nl blk.38.attn_output.weight iq3_s blk.38.attn_v.weight q4_K -blk.39.ffn_down_exps.weight iq4_nl -blk.39.ffn_down_shexp.weight iq4_nl +blk.38.ffn_down_exps.weight iq4_nl +blk.38.ffn_down_shexp.weight iq4_nl blk.39.attn_output.weight iq3_s blk.39.attn_v.weight q4_K -blk.40.ffn_down_exps.weight iq4_nl -blk.40.ffn_down_shexp.weight iq4_nl +blk.39.ffn_down_exps.weight iq4_nl +blk.39.ffn_down_shexp.weight iq4_nl blk.40.attn_output.weight iq3_s blk.40.attn_v.weight q4_K -blk.41.ffn_down_exps.weight iq4_nl -blk.41.ffn_down_shexp.weight iq4_nl +blk.40.ffn_down_exps.weight iq4_nl +blk.40.ffn_down_shexp.weight iq4_nl blk.41.attn_output.weight iq3_s blk.41.attn_v.weight q4_K +blk.41.ffn_down_exps.weight iq4_nl +blk.41.ffn_down_shexp.weight iq4_nl +blk.42.attn_output.weight iq3_s +blk.42.attn_v.weight q4_K blk.42.ffn_down_exps.weight iq4_nl blk.42.ffn_down_shexp.weight iq4_nl -blk.42.attn_v.weight q4_K -blk.42.attn_output.weight iq3_s -blk.43.ffn_down_exps.weight iq4_nl -blk.43.ffn_down_shexp.weight iq4_nl blk.43.attn_output.weight iq3_s blk.43.attn_v.weight q4_K -blk.44.ffn_down_exps.weight iq4_nl -blk.44.ffn_down_shexp.weight iq4_nl +blk.43.ffn_down_exps.weight iq4_nl +blk.43.ffn_down_shexp.weight iq4_nl blk.44.attn_output.weight iq3_s blk.44.attn_v.weight q4_K -output.weight q5_K -blk.45.ffn_down_exps.weight iq4_nl -blk.45.ffn_down_shexp.weight iq4_nl +blk.44.ffn_down_exps.weight iq4_nl +blk.44.ffn_down_shexp.weight iq4_nl blk.45.attn_output.weight iq3_s blk.45.attn_v.weight q4_K +blk.45.ffn_down_exps.weight iq4_nl +blk.45.ffn_down_shexp.weight iq4_nl [IQ2_M] iq2_s +output.weight q5_K token_embd.weight iq3_s -blk.0.ffn_down.weight iq4_nl blk.0.attn_output.weight iq3_s blk.0.attn_v.weight q4_K -blk.1.ffn_down_exps.weight iq4_nl -blk.1.ffn_down_shexp.weight iq4_nl +blk.0.ffn_down.weight iq4_nl blk.1.attn_output.weight iq3_s blk.1.attn_v.weight q4_K -blk.2.ffn_down_exps.weight iq4_nl -blk.2.ffn_down_shexp.weight iq4_nl +blk.1.ffn_down_exps.weight iq4_nl +blk.1.ffn_down_shexp.weight iq4_nl blk.2.attn_output.weight iq3_s blk.2.attn_v.weight q4_K -blk.3.ffn_down_exps.weight iq4_nl -blk.3.ffn_down_shexp.weight iq4_nl +blk.2.ffn_down_exps.weight iq4_nl +blk.2.ffn_down_shexp.weight iq4_nl blk.3.attn_output.weight iq3_s blk.3.attn_v.weight q4_K -blk.4.ffn_down_exps.weight iq4_nl -blk.4.ffn_down_shexp.weight iq4_nl -blk.4.attn_output.weight iq3_s -blk.4.attn_v.weight q4_K -blk.5.ffn_down_exps.weight iq4_nl -blk.5.ffn_down_shexp.weight iq4_nl -blk.5.attn_output.weight iq3_s -blk.5.attn_v.weight q4_K -blk.6.ffn_down_exps.weight iq4_nl -blk.6.ffn_down_shexp.weight iq4_nl -blk.6.attn_output.weight iq3_s -blk.6.attn_v.weight q4_K -blk.7.ffn_down_exps.weight iq4_nl -blk.7.ffn_down_shexp.weight iq4_nl -blk.7.attn_output.weight iq3_s -blk.7.attn_v.weight q4_K -blk.8.ffn_down_exps.weight iq4_nl -blk.8.ffn_down_shexp.weight iq4_nl -blk.8.attn_output.weight iq3_s -blk.8.attn_v.weight q4_K -blk.9.ffn_down_exps.weight iq4_nl -blk.9.ffn_down_shexp.weight iq4_nl -blk.9.attn_output.weight iq3_s -blk.9.attn_v.weight q4_K -blk.10.ffn_down_exps.weight iq4_nl -blk.10.ffn_down_shexp.weight iq4_nl -blk.10.attn_output.weight iq3_s -blk.10.attn_v.weight q4_K -blk.11.ffn_down_exps.weight iq4_nl -blk.11.ffn_down_shexp.weight iq4_nl -blk.11.attn_output.weight iq3_s -blk.11.attn_v.weight q4_K -blk.12.ffn_down_exps.weight iq4_nl -blk.12.ffn_down_shexp.weight iq4_nl -blk.12.attn_output.weight iq3_s -blk.12.attn_v.weight q4_K -blk.13.ffn_down_exps.weight iq4_nl -blk.13.ffn_down_shexp.weight iq4_nl -blk.13.attn_output.weight iq3_s -blk.13.attn_v.weight q4_K -blk.14.ffn_down_exps.weight iq4_nl -blk.14.ffn_down_shexp.weight iq4_nl -blk.14.attn_output.weight iq3_s -blk.14.attn_v.weight q4_K -blk.15.ffn_down_exps.weight iq4_nl -blk.15.ffn_down_shexp.weight iq4_nl -blk.15.attn_output.weight iq3_s -blk.15.attn_v.weight q4_K -blk.16.ffn_down_exps.weight iq4_nl -blk.16.ffn_down_shexp.weight iq4_nl -blk.16.attn_output.weight iq3_s -blk.16.attn_v.weight q4_K -blk.17.ffn_down_exps.weight iq4_nl -blk.17.ffn_down_shexp.weight iq4_nl -blk.17.attn_output.weight iq3_s -blk.17.attn_v.weight q4_K -blk.18.ffn_down_exps.weight iq4_nl -blk.18.ffn_down_shexp.weight iq4_nl -blk.18.attn_output.weight iq3_s -blk.18.attn_v.weight q4_K -blk.19.ffn_down_exps.weight iq4_nl -blk.19.ffn_down_shexp.weight iq4_nl -blk.19.attn_output.weight iq3_s -blk.19.attn_v.weight q4_K -blk.20.ffn_down_exps.weight iq4_nl -blk.20.ffn_down_shexp.weight iq4_nl -blk.20.attn_output.weight iq3_s -blk.20.attn_v.weight q4_K -blk.21.ffn_down_exps.weight iq4_nl -blk.21.ffn_down_shexp.weight iq4_nl -blk.21.attn_output.weight iq3_s -blk.21.attn_v.weight q4_K -blk.22.ffn_down_exps.weight iq4_nl -blk.22.ffn_down_shexp.weight iq4_nl -blk.22.attn_output.weight iq3_s -blk.22.attn_v.weight q4_K -blk.23.ffn_down_exps.weight iq4_nl -blk.23.ffn_down_shexp.weight iq4_nl -blk.23.attn_output.weight iq3_s -blk.23.attn_v.weight q4_K -blk.24.ffn_down_exps.weight iq4_nl -blk.24.ffn_down_shexp.weight iq4_nl -blk.24.attn_output.weight iq3_s -blk.24.attn_v.weight q4_K -blk.25.ffn_down_exps.weight iq4_nl -blk.25.ffn_down_shexp.weight iq4_nl -blk.25.attn_output.weight iq3_s -blk.25.attn_v.weight q4_K -blk.26.ffn_down_exps.weight iq4_nl -blk.26.ffn_down_shexp.weight iq4_nl -blk.26.attn_output.weight iq3_s -blk.26.attn_v.weight q4_K -blk.27.ffn_down_exps.weight iq4_nl -blk.27.ffn_down_shexp.weight iq4_nl -blk.27.attn_output.weight iq3_s -blk.27.attn_v.weight q4_K -blk.28.ffn_down_exps.weight iq4_nl -blk.28.ffn_down_shexp.weight iq4_nl -blk.28.attn_output.weight iq3_s -blk.28.attn_v.weight q4_K -blk.29.ffn_down_exps.weight iq4_nl -blk.29.ffn_down_shexp.weight iq4_nl -blk.29.attn_output.weight iq3_s -blk.29.attn_v.weight q4_K -blk.30.ffn_down_exps.weight iq4_nl -blk.30.ffn_down_shexp.weight iq4_nl -blk.30.attn_output.weight iq3_s -blk.30.attn_v.weight q4_K -blk.31.ffn_down_exps.weight iq4_nl -blk.31.ffn_down_shexp.weight iq4_nl -blk.31.attn_output.weight iq3_s -blk.31.attn_v.weight q4_K -blk.32.ffn_down_exps.weight iq4_nl -blk.32.ffn_down_shexp.weight iq4_nl -blk.32.attn_output.weight iq3_s -blk.32.attn_v.weight q4_K -blk.33.ffn_down_exps.weight iq4_nl -blk.33.ffn_down_shexp.weight iq4_nl -blk.33.attn_output.weight iq3_s -blk.33.attn_v.weight q4_K -blk.34.ffn_down_exps.weight iq4_nl -blk.34.attn_output.weight iq3_s -blk.34.ffn_down_shexp.weight iq4_nl -blk.34.attn_v.weight q4_K -blk.35.ffn_down_exps.weight iq4_nl -blk.35.ffn_down_shexp.weight iq4_nl -blk.35.attn_output.weight iq3_s -blk.35.attn_v.weight q4_K -blk.36.ffn_down_exps.weight iq4_nl -blk.36.ffn_down_shexp.weight iq4_nl -blk.36.attn_output.weight iq3_s -blk.36.attn_v.weight q4_K -blk.37.ffn_down_exps.weight iq4_nl -blk.37.ffn_down_shexp.weight iq4_nl -blk.37.attn_output.weight iq3_s -blk.37.attn_v.weight q4_K -blk.38.ffn_down_exps.weight iq4_nl -blk.38.ffn_down_shexp.weight iq4_nl -blk.38.attn_output.weight iq3_s -blk.38.attn_v.weight q4_K -blk.39.ffn_down_exps.weight iq4_nl -blk.39.ffn_down_shexp.weight iq4_nl -blk.39.attn_output.weight iq3_s -blk.39.attn_v.weight q4_K -blk.40.ffn_down_exps.weight iq4_nl -blk.40.ffn_down_shexp.weight iq4_nl -blk.40.attn_output.weight iq3_s -blk.40.attn_v.weight q4_K -blk.41.ffn_down_exps.weight iq4_nl -blk.41.ffn_down_shexp.weight iq4_nl -blk.41.attn_output.weight iq3_s -blk.41.attn_v.weight q4_K -blk.42.ffn_down_exps.weight iq4_nl -blk.42.ffn_down_shexp.weight iq4_nl -blk.42.attn_v.weight q4_K -blk.42.attn_output.weight iq3_s -blk.43.ffn_down_exps.weight iq4_nl -blk.43.ffn_down_shexp.weight iq4_nl -blk.43.attn_output.weight iq3_s -blk.43.attn_v.weight q4_K -blk.44.ffn_down_exps.weight iq4_nl -blk.44.ffn_down_shexp.weight iq4_nl -blk.44.attn_output.weight iq3_s -blk.44.attn_v.weight q4_K -output.weight q5_K -blk.45.ffn_down_exps.weight iq4_nl -blk.45.ffn_down_shexp.weight iq4_nl -blk.45.attn_output.weight iq3_s -blk.45.attn_v.weight q4_K - -[IQ4_XS] iq4_xs -blk.0.ffn_down.weight q5_1 -blk.0.attn_v.weight q5_K -blk.1.ffn_down_exps.weight q5_1 -blk.1.ffn_down_shexp.weight q5_1 -blk.1.attn_v.weight q5_K -blk.2.ffn_down_exps.weight q5_1 -blk.2.ffn_down_shexp.weight q5_1 -blk.2.attn_v.weight q5_K -blk.3.ffn_down_exps.weight q5_1 -blk.3.ffn_down_shexp.weight q5_1 -blk.3.attn_v.weight q5_K -blk.4.ffn_down_exps.weight q5_1 -blk.4.ffn_down_shexp.weight q5_1 -blk.4.attn_v.weight q5_K -blk.5.ffn_down_exps.weight iq4_nl -blk.5.ffn_down_shexp.weight iq4_nl -blk.5.attn_v.weight q5_K -blk.6.ffn_down_exps.weight iq4_nl -blk.6.ffn_down_shexp.weight iq4_nl -blk.6.attn_v.weight q5_K -blk.7.ffn_down_exps.weight iq4_nl -blk.7.ffn_down_shexp.weight iq4_nl -blk.7.attn_v.weight q5_K -blk.8.ffn_down_exps.weight iq4_nl -blk.8.ffn_down_shexp.weight iq4_nl -blk.8.attn_v.weight q5_K -blk.9.ffn_down_exps.weight iq4_nl -blk.9.ffn_down_shexp.weight iq4_nl -blk.9.attn_v.weight q5_K -blk.10.ffn_down_exps.weight iq4_nl -blk.10.ffn_down_shexp.weight iq4_nl -blk.10.attn_v.weight q5_K -blk.11.ffn_down_exps.weight iq4_nl -blk.11.ffn_down_shexp.weight iq4_nl -blk.11.attn_v.weight q5_K -blk.12.ffn_down_exps.weight iq4_nl -blk.12.ffn_down_shexp.weight iq4_nl -blk.12.attn_v.weight q5_K -blk.13.ffn_down_exps.weight iq4_nl -blk.13.ffn_down_shexp.weight iq4_nl -blk.13.attn_v.weight q5_K -blk.14.ffn_down_exps.weight iq4_nl -blk.14.ffn_down_shexp.weight iq4_nl -blk.14.attn_v.weight q5_K -blk.15.ffn_down_exps.weight iq4_nl -blk.15.ffn_down_shexp.weight iq4_nl -blk.15.attn_v.weight q5_K -blk.16.ffn_down_exps.weight iq4_nl -blk.16.ffn_down_shexp.weight iq4_nl -blk.16.attn_v.weight q5_K -blk.17.ffn_down_exps.weight iq4_nl -blk.17.ffn_down_shexp.weight iq4_nl -blk.17.attn_v.weight q5_K -blk.18.ffn_down_exps.weight iq4_nl -blk.18.ffn_down_shexp.weight iq4_nl -blk.18.attn_v.weight q5_K -blk.19.ffn_down_exps.weight iq4_nl -blk.19.ffn_down_shexp.weight iq4_nl -blk.19.attn_v.weight q5_K -blk.20.ffn_down_exps.weight iq4_nl -blk.20.ffn_down_shexp.weight iq4_nl -blk.20.attn_v.weight q5_K -blk.21.ffn_down_exps.weight iq4_nl -blk.21.ffn_down_shexp.weight iq4_nl -blk.21.attn_v.weight q5_K -blk.22.ffn_down_exps.weight iq4_nl -blk.22.ffn_down_shexp.weight iq4_nl -blk.22.attn_v.weight q5_K -blk.23.ffn_down_exps.weight iq4_nl -blk.23.ffn_down_shexp.weight iq4_nl -blk.23.attn_v.weight q5_K -blk.24.ffn_down_exps.weight iq4_nl -blk.24.ffn_down_shexp.weight iq4_nl -blk.24.attn_v.weight q5_K -blk.25.ffn_down_exps.weight iq4_nl -blk.25.ffn_down_shexp.weight iq4_nl -blk.25.attn_v.weight q5_K -blk.26.ffn_down_exps.weight iq4_nl -blk.26.ffn_down_shexp.weight iq4_nl -blk.26.attn_v.weight q5_K -blk.27.ffn_down_exps.weight iq4_nl -blk.27.ffn_down_shexp.weight iq4_nl -blk.27.attn_v.weight q5_K -blk.28.ffn_down_exps.weight iq4_nl -blk.28.ffn_down_shexp.weight iq4_nl -blk.28.attn_v.weight q5_K -blk.29.ffn_down_exps.weight iq4_nl -blk.29.ffn_down_shexp.weight iq4_nl -blk.29.attn_v.weight q5_K -blk.30.ffn_down_exps.weight iq4_nl -blk.30.ffn_down_shexp.weight iq4_nl -blk.30.attn_v.weight q5_K -blk.31.ffn_down_exps.weight iq4_nl -blk.31.ffn_down_shexp.weight iq4_nl -blk.31.attn_v.weight q5_K -blk.32.ffn_down_exps.weight iq4_nl -blk.32.ffn_down_shexp.weight iq4_nl -blk.32.attn_v.weight q5_K -blk.33.ffn_down_exps.weight iq4_nl -blk.33.ffn_down_shexp.weight iq4_nl -blk.33.attn_v.weight q5_K -blk.34.ffn_down_exps.weight iq4_nl -blk.34.ffn_down_shexp.weight iq4_nl -blk.34.attn_v.weight q5_K -blk.35.ffn_down_exps.weight iq4_nl -blk.35.ffn_down_shexp.weight iq4_nl -blk.35.attn_v.weight q5_K -blk.36.ffn_down_exps.weight iq4_nl -blk.36.ffn_down_shexp.weight iq4_nl -blk.36.attn_v.weight q5_K -blk.37.ffn_down_exps.weight iq4_nl -blk.37.ffn_down_shexp.weight iq4_nl -blk.37.attn_v.weight q5_K -blk.38.ffn_down_exps.weight iq4_nl -blk.38.ffn_down_shexp.weight iq4_nl -blk.38.attn_v.weight q5_K -blk.39.ffn_down_exps.weight iq4_nl -blk.39.ffn_down_shexp.weight iq4_nl -blk.39.attn_v.weight q5_K -blk.40.ffn_down_exps.weight iq4_nl -blk.40.ffn_down_shexp.weight iq4_nl -blk.40.attn_v.weight q5_K -blk.41.ffn_down_exps.weight iq4_nl -blk.41.ffn_down_shexp.weight iq4_nl -blk.41.attn_v.weight q5_K -blk.42.ffn_down_exps.weight iq4_nl -blk.42.ffn_down_shexp.weight iq4_nl -blk.42.attn_v.weight q5_K -blk.43.ffn_down_exps.weight iq4_nl -blk.43.ffn_down_shexp.weight iq4_nl -blk.43.attn_v.weight q5_K -blk.44.ffn_down_exps.weight iq4_nl -blk.44.ffn_down_shexp.weight iq4_nl -blk.44.attn_v.weight q5_K -output.weight q6_K -blk.45.ffn_down_exps.weight iq4_nl -blk.45.ffn_down_shexp.weight iq4_nl -blk.45.attn_v.weight q5_K - -[IQ1_M] iq1_m -token_embd.weight q2_K -blk.0.ffn_down.weight q4_0 -blk.0.attn_output.weight iq2_xxs -blk.0.attn_v.weight q4_K -blk.1.ffn_down_exps.weight q4_0 -blk.1.ffn_down_shexp.weight q4_0 -blk.1.attn_output.weight iq2_xxs -blk.1.attn_v.weight q4_K -blk.2.ffn_down_exps.weight q4_0 -blk.2.ffn_down_shexp.weight q4_0 -blk.2.attn_output.weight iq2_xxs -blk.2.attn_v.weight q4_K blk.3.ffn_down_exps.weight iq4_nl blk.3.ffn_down_shexp.weight iq4_nl -blk.3.attn_output.weight iq2_xxs -blk.3.attn_v.weight q4_K +blk.4.attn_output.weight iq3_s +blk.4.attn_v.weight q4_K blk.4.ffn_down_exps.weight iq4_nl blk.4.ffn_down_shexp.weight iq4_nl -blk.4.attn_output.weight iq2_xxs -blk.4.attn_v.weight q4_K +blk.5.attn_output.weight iq3_s +blk.5.attn_v.weight q4_K blk.5.ffn_down_exps.weight iq4_nl blk.5.ffn_down_shexp.weight iq4_nl -blk.5.attn_output.weight iq2_xxs -blk.5.attn_v.weight q4_K +blk.6.attn_output.weight iq3_s +blk.6.attn_v.weight q4_K blk.6.ffn_down_exps.weight iq4_nl blk.6.ffn_down_shexp.weight iq4_nl -blk.6.attn_output.weight iq2_xxs -blk.6.attn_v.weight q4_K +blk.7.attn_output.weight iq3_s +blk.7.attn_v.weight q4_K blk.7.ffn_down_exps.weight iq4_nl blk.7.ffn_down_shexp.weight iq4_nl -blk.7.attn_output.weight iq2_xxs -blk.7.attn_v.weight q4_K +blk.8.attn_output.weight iq3_s +blk.8.attn_v.weight q4_K blk.8.ffn_down_exps.weight iq4_nl blk.8.ffn_down_shexp.weight iq4_nl -blk.8.attn_output.weight iq2_xxs -blk.8.attn_v.weight q4_K +blk.9.attn_output.weight iq3_s +blk.9.attn_v.weight q4_K blk.9.ffn_down_exps.weight iq4_nl blk.9.ffn_down_shexp.weight iq4_nl -blk.9.attn_output.weight iq2_xxs -blk.9.attn_v.weight q4_K +blk.10.attn_output.weight iq3_s +blk.10.attn_v.weight q4_K blk.10.ffn_down_exps.weight iq4_nl blk.10.ffn_down_shexp.weight iq4_nl -blk.10.attn_output.weight iq2_xxs -blk.10.attn_v.weight q4_K +blk.11.attn_output.weight iq3_s +blk.11.attn_v.weight q4_K blk.11.ffn_down_exps.weight iq4_nl blk.11.ffn_down_shexp.weight iq4_nl -blk.11.attn_output.weight iq2_xxs -blk.11.attn_v.weight q4_K +blk.12.attn_output.weight iq3_s +blk.12.attn_v.weight q4_K blk.12.ffn_down_exps.weight iq4_nl blk.12.ffn_down_shexp.weight iq4_nl -blk.12.attn_output.weight iq2_xxs -blk.12.attn_v.weight q4_K +blk.13.attn_output.weight iq3_s +blk.13.attn_v.weight q4_K blk.13.ffn_down_exps.weight iq4_nl blk.13.ffn_down_shexp.weight iq4_nl -blk.13.attn_output.weight iq2_xxs -blk.13.attn_v.weight q4_K +blk.14.attn_output.weight iq3_s +blk.14.attn_v.weight q4_K blk.14.ffn_down_exps.weight iq4_nl blk.14.ffn_down_shexp.weight iq4_nl -blk.14.attn_output.weight iq2_xxs -blk.14.attn_v.weight q4_K +blk.15.attn_output.weight iq3_s +blk.15.attn_v.weight q4_K blk.15.ffn_down_exps.weight iq4_nl blk.15.ffn_down_shexp.weight iq4_nl -blk.15.attn_output.weight iq2_xxs -blk.15.attn_v.weight q4_K +blk.16.attn_output.weight iq3_s +blk.16.attn_v.weight q4_K blk.16.ffn_down_exps.weight iq4_nl blk.16.ffn_down_shexp.weight iq4_nl -blk.16.attn_output.weight iq2_xxs -blk.16.attn_v.weight q4_K +blk.17.attn_output.weight iq3_s +blk.17.attn_v.weight q4_K blk.17.ffn_down_exps.weight iq4_nl blk.17.ffn_down_shexp.weight iq4_nl -blk.17.attn_output.weight iq2_xxs -blk.17.attn_v.weight q4_K +blk.18.attn_output.weight iq3_s +blk.18.attn_v.weight q4_K blk.18.ffn_down_exps.weight iq4_nl blk.18.ffn_down_shexp.weight iq4_nl -blk.18.attn_output.weight iq2_xxs -blk.18.attn_v.weight q4_K +blk.19.attn_output.weight iq3_s +blk.19.attn_v.weight q4_K blk.19.ffn_down_exps.weight iq4_nl blk.19.ffn_down_shexp.weight iq4_nl -blk.19.attn_output.weight iq2_xxs -blk.19.attn_v.weight q4_K +blk.20.attn_output.weight iq3_s +blk.20.attn_v.weight q4_K blk.20.ffn_down_exps.weight iq4_nl blk.20.ffn_down_shexp.weight iq4_nl -blk.20.attn_output.weight iq2_xxs -blk.20.attn_v.weight q4_K +blk.21.attn_output.weight iq3_s +blk.21.attn_v.weight q4_K blk.21.ffn_down_exps.weight iq4_nl blk.21.ffn_down_shexp.weight iq4_nl -blk.21.attn_output.weight iq2_xxs -blk.21.attn_v.weight q4_K +blk.22.attn_output.weight iq3_s +blk.22.attn_v.weight q4_K blk.22.ffn_down_exps.weight iq4_nl blk.22.ffn_down_shexp.weight iq4_nl -blk.22.attn_output.weight iq2_xxs -blk.22.attn_v.weight q4_K +blk.23.attn_output.weight iq3_s +blk.23.attn_v.weight q4_K blk.23.ffn_down_exps.weight iq4_nl blk.23.ffn_down_shexp.weight iq4_nl -blk.23.attn_output.weight iq2_xxs -blk.23.attn_v.weight q4_K +blk.24.attn_output.weight iq3_s +blk.24.attn_v.weight q4_K blk.24.ffn_down_exps.weight iq4_nl blk.24.ffn_down_shexp.weight iq4_nl -blk.24.attn_output.weight iq2_xxs -blk.24.attn_v.weight q4_K +blk.25.attn_output.weight iq3_s +blk.25.attn_v.weight q4_K blk.25.ffn_down_exps.weight iq4_nl blk.25.ffn_down_shexp.weight iq4_nl -blk.25.attn_output.weight iq2_xxs -blk.25.attn_v.weight q4_K +blk.26.attn_output.weight iq3_s +blk.26.attn_v.weight q4_K blk.26.ffn_down_exps.weight iq4_nl blk.26.ffn_down_shexp.weight iq4_nl -blk.26.attn_output.weight iq2_xxs -blk.26.attn_v.weight q4_K +blk.27.attn_output.weight iq3_s +blk.27.attn_v.weight q4_K blk.27.ffn_down_exps.weight iq4_nl blk.27.ffn_down_shexp.weight iq4_nl -blk.27.attn_output.weight iq2_xxs -blk.27.attn_v.weight q4_K +blk.28.attn_output.weight iq3_s +blk.28.attn_v.weight q4_K blk.28.ffn_down_exps.weight iq4_nl blk.28.ffn_down_shexp.weight iq4_nl -blk.28.attn_output.weight iq2_xxs -blk.28.attn_v.weight q4_K +blk.29.attn_output.weight iq3_s +blk.29.attn_v.weight q4_K blk.29.ffn_down_exps.weight iq4_nl blk.29.ffn_down_shexp.weight iq4_nl -blk.29.attn_output.weight iq2_xxs -blk.29.attn_v.weight q4_K +blk.30.attn_output.weight iq3_s +blk.30.attn_v.weight q4_K blk.30.ffn_down_exps.weight iq4_nl blk.30.ffn_down_shexp.weight iq4_nl -blk.30.attn_output.weight iq2_xxs -blk.30.attn_v.weight q4_K +blk.31.attn_output.weight iq3_s +blk.31.attn_v.weight q4_K blk.31.ffn_down_exps.weight iq4_nl blk.31.ffn_down_shexp.weight iq4_nl -blk.31.attn_output.weight iq2_xxs -blk.31.attn_v.weight q4_K +blk.32.attn_output.weight iq3_s +blk.32.attn_v.weight q4_K blk.32.ffn_down_exps.weight iq4_nl blk.32.ffn_down_shexp.weight iq4_nl -blk.32.attn_output.weight iq2_xxs -blk.32.attn_v.weight q4_K +blk.33.attn_output.weight iq3_s +blk.33.attn_v.weight q4_K blk.33.ffn_down_exps.weight iq4_nl blk.33.ffn_down_shexp.weight iq4_nl -blk.33.attn_output.weight iq2_xxs -blk.33.attn_v.weight q4_K -blk.34.ffn_down_exps.weight iq4_nl -blk.34.attn_output.weight iq2_xxs -blk.34.ffn_down_shexp.weight iq4_nl +blk.34.attn_output.weight iq3_s blk.34.attn_v.weight q4_K +blk.34.ffn_down_exps.weight iq4_nl +blk.34.ffn_down_shexp.weight iq4_nl +blk.35.attn_output.weight iq3_s +blk.35.attn_v.weight q4_K blk.35.ffn_down_exps.weight iq4_nl blk.35.ffn_down_shexp.weight iq4_nl -blk.35.attn_output.weight iq2_xxs -blk.35.attn_v.weight q4_K +blk.36.attn_output.weight iq3_s +blk.36.attn_v.weight q4_K blk.36.ffn_down_exps.weight iq4_nl blk.36.ffn_down_shexp.weight iq4_nl -blk.36.attn_output.weight iq2_xxs -blk.36.attn_v.weight q4_K +blk.37.attn_output.weight iq3_s +blk.37.attn_v.weight q4_K blk.37.ffn_down_exps.weight iq4_nl blk.37.ffn_down_shexp.weight iq4_nl -blk.37.attn_output.weight iq2_xxs -blk.37.attn_v.weight q4_K +blk.38.attn_output.weight iq3_s +blk.38.attn_v.weight q4_K blk.38.ffn_down_exps.weight iq4_nl blk.38.ffn_down_shexp.weight iq4_nl -blk.38.attn_output.weight iq2_xxs -blk.38.attn_v.weight q4_K +blk.39.attn_output.weight iq3_s +blk.39.attn_v.weight q4_K blk.39.ffn_down_exps.weight iq4_nl blk.39.ffn_down_shexp.weight iq4_nl -blk.39.attn_output.weight iq2_xxs -blk.39.attn_v.weight q4_K +blk.40.attn_output.weight iq3_s +blk.40.attn_v.weight q4_K blk.40.ffn_down_exps.weight iq4_nl blk.40.ffn_down_shexp.weight iq4_nl -blk.40.attn_output.weight iq2_xxs -blk.40.attn_v.weight q4_K +blk.41.attn_output.weight iq3_s +blk.41.attn_v.weight q4_K blk.41.ffn_down_exps.weight iq4_nl blk.41.ffn_down_shexp.weight iq4_nl -blk.41.attn_output.weight iq2_xxs -blk.41.attn_v.weight q4_K +blk.42.attn_output.weight iq3_s +blk.42.attn_v.weight q4_K blk.42.ffn_down_exps.weight iq4_nl blk.42.ffn_down_shexp.weight iq4_nl -blk.42.attn_v.weight q4_K -blk.42.attn_output.weight iq2_xxs +blk.43.attn_output.weight iq3_s +blk.43.attn_v.weight q4_K blk.43.ffn_down_exps.weight iq4_nl blk.43.ffn_down_shexp.weight iq4_nl -blk.43.attn_output.weight iq2_xxs -blk.43.attn_v.weight q4_K +blk.44.attn_output.weight iq3_s +blk.44.attn_v.weight q4_K blk.44.ffn_down_exps.weight iq4_nl blk.44.ffn_down_shexp.weight iq4_nl -blk.44.attn_output.weight iq2_xxs -blk.44.attn_v.weight q4_K -output.weight q5_K +blk.45.attn_output.weight iq3_s +blk.45.attn_v.weight q4_K blk.45.ffn_down_exps.weight iq4_nl blk.45.ffn_down_shexp.weight iq4_nl + +[IQ4_XS] iq4_xs +output.weight q6_K +blk.0.attn_v.weight q5_K +blk.0.ffn_down.weight q5_1 +blk.1.attn_v.weight q5_K +blk.1.ffn_down_exps.weight q5_1 +blk.1.ffn_down_shexp.weight q5_1 +blk.2.attn_v.weight q5_K +blk.2.ffn_down_exps.weight q5_1 +blk.2.ffn_down_shexp.weight q5_1 +blk.3.attn_v.weight q5_K +blk.3.ffn_down_exps.weight q5_1 +blk.3.ffn_down_shexp.weight q5_1 +blk.4.attn_v.weight q5_K +blk.4.ffn_down_exps.weight q5_1 +blk.4.ffn_down_shexp.weight q5_1 +blk.5.attn_v.weight q5_K +blk.5.ffn_down_exps.weight iq4_nl +blk.5.ffn_down_shexp.weight iq4_nl +blk.6.attn_v.weight q5_K +blk.6.ffn_down_exps.weight iq4_nl +blk.6.ffn_down_shexp.weight iq4_nl +blk.7.attn_v.weight q5_K +blk.7.ffn_down_exps.weight iq4_nl +blk.7.ffn_down_shexp.weight iq4_nl +blk.8.attn_v.weight q5_K +blk.8.ffn_down_exps.weight iq4_nl +blk.8.ffn_down_shexp.weight iq4_nl +blk.9.attn_v.weight q5_K +blk.9.ffn_down_exps.weight iq4_nl +blk.9.ffn_down_shexp.weight iq4_nl +blk.10.attn_v.weight q5_K +blk.10.ffn_down_exps.weight iq4_nl +blk.10.ffn_down_shexp.weight iq4_nl +blk.11.attn_v.weight q5_K +blk.11.ffn_down_exps.weight iq4_nl +blk.11.ffn_down_shexp.weight iq4_nl +blk.12.attn_v.weight q5_K +blk.12.ffn_down_exps.weight iq4_nl +blk.12.ffn_down_shexp.weight iq4_nl +blk.13.attn_v.weight q5_K +blk.13.ffn_down_exps.weight iq4_nl +blk.13.ffn_down_shexp.weight iq4_nl +blk.14.attn_v.weight q5_K +blk.14.ffn_down_exps.weight iq4_nl +blk.14.ffn_down_shexp.weight iq4_nl +blk.15.attn_v.weight q5_K +blk.15.ffn_down_exps.weight iq4_nl +blk.15.ffn_down_shexp.weight iq4_nl +blk.16.attn_v.weight q5_K +blk.16.ffn_down_exps.weight iq4_nl +blk.16.ffn_down_shexp.weight iq4_nl +blk.17.attn_v.weight q5_K +blk.17.ffn_down_exps.weight iq4_nl +blk.17.ffn_down_shexp.weight iq4_nl +blk.18.attn_v.weight q5_K +blk.18.ffn_down_exps.weight iq4_nl +blk.18.ffn_down_shexp.weight iq4_nl +blk.19.attn_v.weight q5_K +blk.19.ffn_down_exps.weight iq4_nl +blk.19.ffn_down_shexp.weight iq4_nl +blk.20.attn_v.weight q5_K +blk.20.ffn_down_exps.weight iq4_nl +blk.20.ffn_down_shexp.weight iq4_nl +blk.21.attn_v.weight q5_K +blk.21.ffn_down_exps.weight iq4_nl +blk.21.ffn_down_shexp.weight iq4_nl +blk.22.attn_v.weight q5_K +blk.22.ffn_down_exps.weight iq4_nl +blk.22.ffn_down_shexp.weight iq4_nl +blk.23.attn_v.weight q5_K +blk.23.ffn_down_exps.weight iq4_nl +blk.23.ffn_down_shexp.weight iq4_nl +blk.24.attn_v.weight q5_K +blk.24.ffn_down_exps.weight iq4_nl +blk.24.ffn_down_shexp.weight iq4_nl +blk.25.attn_v.weight q5_K +blk.25.ffn_down_exps.weight iq4_nl +blk.25.ffn_down_shexp.weight iq4_nl +blk.26.attn_v.weight q5_K +blk.26.ffn_down_exps.weight iq4_nl +blk.26.ffn_down_shexp.weight iq4_nl +blk.27.attn_v.weight q5_K +blk.27.ffn_down_exps.weight iq4_nl +blk.27.ffn_down_shexp.weight iq4_nl +blk.28.attn_v.weight q5_K +blk.28.ffn_down_exps.weight iq4_nl +blk.28.ffn_down_shexp.weight iq4_nl +blk.29.attn_v.weight q5_K +blk.29.ffn_down_exps.weight iq4_nl +blk.29.ffn_down_shexp.weight iq4_nl +blk.30.attn_v.weight q5_K +blk.30.ffn_down_exps.weight iq4_nl +blk.30.ffn_down_shexp.weight iq4_nl +blk.31.attn_v.weight q5_K +blk.31.ffn_down_exps.weight iq4_nl +blk.31.ffn_down_shexp.weight iq4_nl +blk.32.attn_v.weight q5_K +blk.32.ffn_down_exps.weight iq4_nl +blk.32.ffn_down_shexp.weight iq4_nl +blk.33.attn_v.weight q5_K +blk.33.ffn_down_exps.weight iq4_nl +blk.33.ffn_down_shexp.weight iq4_nl +blk.34.attn_v.weight q5_K +blk.34.ffn_down_exps.weight iq4_nl +blk.34.ffn_down_shexp.weight iq4_nl +blk.35.attn_v.weight q5_K +blk.35.ffn_down_exps.weight iq4_nl +blk.35.ffn_down_shexp.weight iq4_nl +blk.36.attn_v.weight q5_K +blk.36.ffn_down_exps.weight iq4_nl +blk.36.ffn_down_shexp.weight iq4_nl +blk.37.attn_v.weight q5_K +blk.37.ffn_down_exps.weight iq4_nl +blk.37.ffn_down_shexp.weight iq4_nl +blk.38.attn_v.weight q5_K +blk.38.ffn_down_exps.weight iq4_nl +blk.38.ffn_down_shexp.weight iq4_nl +blk.39.attn_v.weight q5_K +blk.39.ffn_down_exps.weight iq4_nl +blk.39.ffn_down_shexp.weight iq4_nl +blk.40.attn_v.weight q5_K +blk.40.ffn_down_exps.weight iq4_nl +blk.40.ffn_down_shexp.weight iq4_nl +blk.41.attn_v.weight q5_K +blk.41.ffn_down_exps.weight iq4_nl +blk.41.ffn_down_shexp.weight iq4_nl +blk.42.attn_v.weight q5_K +blk.42.ffn_down_exps.weight iq4_nl +blk.42.ffn_down_shexp.weight iq4_nl +blk.43.attn_v.weight q5_K +blk.43.ffn_down_exps.weight iq4_nl +blk.43.ffn_down_shexp.weight iq4_nl +blk.44.attn_v.weight q5_K +blk.44.ffn_down_exps.weight iq4_nl +blk.44.ffn_down_shexp.weight iq4_nl +blk.45.attn_v.weight q5_K +blk.45.ffn_down_exps.weight iq4_nl +blk.45.ffn_down_shexp.weight iq4_nl + +[IQ1_M] iq1_m +output.weight q5_K +token_embd.weight q2_K +blk.0.attn_output.weight iq2_xxs +blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q4_0 +blk.1.attn_output.weight iq2_xxs +blk.1.attn_v.weight q4_K +blk.1.ffn_down_exps.weight q4_0 +blk.1.ffn_down_shexp.weight q4_0 +blk.2.attn_output.weight iq2_xxs +blk.2.attn_v.weight q4_K +blk.2.ffn_down_exps.weight q4_0 +blk.2.ffn_down_shexp.weight q4_0 +blk.3.attn_output.weight iq2_xxs +blk.3.attn_v.weight q4_K +blk.3.ffn_down_exps.weight iq4_nl +blk.3.ffn_down_shexp.weight iq4_nl +blk.4.attn_output.weight iq2_xxs +blk.4.attn_v.weight q4_K +blk.4.ffn_down_exps.weight iq4_nl +blk.4.ffn_down_shexp.weight iq4_nl +blk.5.attn_output.weight iq2_xxs +blk.5.attn_v.weight q4_K +blk.5.ffn_down_exps.weight iq4_nl +blk.5.ffn_down_shexp.weight iq4_nl +blk.6.attn_output.weight iq2_xxs +blk.6.attn_v.weight q4_K +blk.6.ffn_down_exps.weight iq4_nl +blk.6.ffn_down_shexp.weight iq4_nl +blk.7.attn_output.weight iq2_xxs +blk.7.attn_v.weight q4_K +blk.7.ffn_down_exps.weight iq4_nl +blk.7.ffn_down_shexp.weight iq4_nl +blk.8.attn_output.weight iq2_xxs +blk.8.attn_v.weight q4_K +blk.8.ffn_down_exps.weight iq4_nl +blk.8.ffn_down_shexp.weight iq4_nl +blk.9.attn_output.weight iq2_xxs +blk.9.attn_v.weight q4_K +blk.9.ffn_down_exps.weight iq4_nl +blk.9.ffn_down_shexp.weight iq4_nl +blk.10.attn_output.weight iq2_xxs +blk.10.attn_v.weight q4_K +blk.10.ffn_down_exps.weight iq4_nl +blk.10.ffn_down_shexp.weight iq4_nl +blk.11.attn_output.weight iq2_xxs +blk.11.attn_v.weight q4_K +blk.11.ffn_down_exps.weight iq4_nl +blk.11.ffn_down_shexp.weight iq4_nl +blk.12.attn_output.weight iq2_xxs +blk.12.attn_v.weight q4_K +blk.12.ffn_down_exps.weight iq4_nl +blk.12.ffn_down_shexp.weight iq4_nl +blk.13.attn_output.weight iq2_xxs +blk.13.attn_v.weight q4_K +blk.13.ffn_down_exps.weight iq4_nl +blk.13.ffn_down_shexp.weight iq4_nl +blk.14.attn_output.weight iq2_xxs +blk.14.attn_v.weight q4_K +blk.14.ffn_down_exps.weight iq4_nl +blk.14.ffn_down_shexp.weight iq4_nl +blk.15.attn_output.weight iq2_xxs +blk.15.attn_v.weight q4_K +blk.15.ffn_down_exps.weight iq4_nl +blk.15.ffn_down_shexp.weight iq4_nl +blk.16.attn_output.weight iq2_xxs +blk.16.attn_v.weight q4_K +blk.16.ffn_down_exps.weight iq4_nl +blk.16.ffn_down_shexp.weight iq4_nl +blk.17.attn_output.weight iq2_xxs +blk.17.attn_v.weight q4_K +blk.17.ffn_down_exps.weight iq4_nl +blk.17.ffn_down_shexp.weight iq4_nl +blk.18.attn_output.weight iq2_xxs +blk.18.attn_v.weight q4_K +blk.18.ffn_down_exps.weight iq4_nl +blk.18.ffn_down_shexp.weight iq4_nl +blk.19.attn_output.weight iq2_xxs +blk.19.attn_v.weight q4_K +blk.19.ffn_down_exps.weight iq4_nl +blk.19.ffn_down_shexp.weight iq4_nl +blk.20.attn_output.weight iq2_xxs +blk.20.attn_v.weight q4_K +blk.20.ffn_down_exps.weight iq4_nl +blk.20.ffn_down_shexp.weight iq4_nl +blk.21.attn_output.weight iq2_xxs +blk.21.attn_v.weight q4_K +blk.21.ffn_down_exps.weight iq4_nl +blk.21.ffn_down_shexp.weight iq4_nl +blk.22.attn_output.weight iq2_xxs +blk.22.attn_v.weight q4_K +blk.22.ffn_down_exps.weight iq4_nl +blk.22.ffn_down_shexp.weight iq4_nl +blk.23.attn_output.weight iq2_xxs +blk.23.attn_v.weight q4_K +blk.23.ffn_down_exps.weight iq4_nl +blk.23.ffn_down_shexp.weight iq4_nl +blk.24.attn_output.weight iq2_xxs +blk.24.attn_v.weight q4_K +blk.24.ffn_down_exps.weight iq4_nl +blk.24.ffn_down_shexp.weight iq4_nl +blk.25.attn_output.weight iq2_xxs +blk.25.attn_v.weight q4_K +blk.25.ffn_down_exps.weight iq4_nl +blk.25.ffn_down_shexp.weight iq4_nl +blk.26.attn_output.weight iq2_xxs +blk.26.attn_v.weight q4_K +blk.26.ffn_down_exps.weight iq4_nl +blk.26.ffn_down_shexp.weight iq4_nl +blk.27.attn_output.weight iq2_xxs +blk.27.attn_v.weight q4_K +blk.27.ffn_down_exps.weight iq4_nl +blk.27.ffn_down_shexp.weight iq4_nl +blk.28.attn_output.weight iq2_xxs +blk.28.attn_v.weight q4_K +blk.28.ffn_down_exps.weight iq4_nl +blk.28.ffn_down_shexp.weight iq4_nl +blk.29.attn_output.weight iq2_xxs +blk.29.attn_v.weight q4_K +blk.29.ffn_down_exps.weight iq4_nl +blk.29.ffn_down_shexp.weight iq4_nl +blk.30.attn_output.weight iq2_xxs +blk.30.attn_v.weight q4_K +blk.30.ffn_down_exps.weight iq4_nl +blk.30.ffn_down_shexp.weight iq4_nl +blk.31.attn_output.weight iq2_xxs +blk.31.attn_v.weight q4_K +blk.31.ffn_down_exps.weight iq4_nl +blk.31.ffn_down_shexp.weight iq4_nl +blk.32.attn_output.weight iq2_xxs +blk.32.attn_v.weight q4_K +blk.32.ffn_down_exps.weight iq4_nl +blk.32.ffn_down_shexp.weight iq4_nl +blk.33.attn_output.weight iq2_xxs +blk.33.attn_v.weight q4_K +blk.33.ffn_down_exps.weight iq4_nl +blk.33.ffn_down_shexp.weight iq4_nl +blk.34.attn_output.weight iq2_xxs +blk.34.attn_v.weight q4_K +blk.34.ffn_down_exps.weight iq4_nl +blk.34.ffn_down_shexp.weight iq4_nl +blk.35.attn_output.weight iq2_xxs +blk.35.attn_v.weight q4_K +blk.35.ffn_down_exps.weight iq4_nl +blk.35.ffn_down_shexp.weight iq4_nl +blk.36.attn_output.weight iq2_xxs +blk.36.attn_v.weight q4_K +blk.36.ffn_down_exps.weight iq4_nl +blk.36.ffn_down_shexp.weight iq4_nl +blk.37.attn_output.weight iq2_xxs +blk.37.attn_v.weight q4_K +blk.37.ffn_down_exps.weight iq4_nl +blk.37.ffn_down_shexp.weight iq4_nl +blk.38.attn_output.weight iq2_xxs +blk.38.attn_v.weight q4_K +blk.38.ffn_down_exps.weight iq4_nl +blk.38.ffn_down_shexp.weight iq4_nl +blk.39.attn_output.weight iq2_xxs +blk.39.attn_v.weight q4_K +blk.39.ffn_down_exps.weight iq4_nl +blk.39.ffn_down_shexp.weight iq4_nl +blk.40.attn_output.weight iq2_xxs +blk.40.attn_v.weight q4_K +blk.40.ffn_down_exps.weight iq4_nl +blk.40.ffn_down_shexp.weight iq4_nl +blk.41.attn_output.weight iq2_xxs +blk.41.attn_v.weight q4_K +blk.41.ffn_down_exps.weight iq4_nl +blk.41.ffn_down_shexp.weight iq4_nl +blk.42.attn_output.weight iq2_xxs +blk.42.attn_v.weight q4_K +blk.42.ffn_down_exps.weight iq4_nl +blk.42.ffn_down_shexp.weight iq4_nl +blk.43.attn_output.weight iq2_xxs +blk.43.attn_v.weight q4_K +blk.43.ffn_down_exps.weight iq4_nl +blk.43.ffn_down_shexp.weight iq4_nl +blk.44.attn_output.weight iq2_xxs +blk.44.attn_v.weight q4_K +blk.44.ffn_down_exps.weight iq4_nl +blk.44.ffn_down_shexp.weight iq4_nl blk.45.attn_output.weight iq2_xxs blk.45.attn_v.weight q4_K +blk.45.ffn_down_exps.weight iq4_nl +blk.45.ffn_down_shexp.weight iq4_nl [BF16] bf16 [TQ1_0] tq1_0 +output.weight q6_K token_embd.weight q4_K blk.0.ffn_down.weight q4_0 blk.1.ffn_down_exps.weight q4_0 @@ -3626,11 +3627,11 @@ blk.43.ffn_down_exps.weight q4_0 blk.43.ffn_down_shexp.weight q4_0 blk.44.ffn_down_exps.weight q4_0 blk.44.ffn_down_shexp.weight q4_0 -output.weight q6_K blk.45.ffn_down_exps.weight q4_0 blk.45.ffn_down_shexp.weight q4_0 [TQ2_0] tq2_0 +output.weight q6_K token_embd.weight q4_K blk.0.ffn_down.weight q4_0 blk.1.ffn_down_exps.weight q4_0 @@ -3721,332 +3722,331 @@ blk.43.ffn_down_exps.weight q4_0 blk.43.ffn_down_shexp.weight q4_0 blk.44.ffn_down_exps.weight q4_0 blk.44.ffn_down_shexp.weight q4_0 -output.weight q6_K blk.45.ffn_down_exps.weight q4_0 blk.45.ffn_down_shexp.weight q4_0 [MXFP4_MOE] mxfp4 +output.weight q8_0 token_embd.weight q8_0 -blk.0.ffn_down.weight q8_0 -blk.0.ffn_gate.weight q8_0 -blk.0.ffn_up.weight q8_0 blk.0.attn_k.weight q8_0 blk.0.attn_output.weight q8_0 blk.0.attn_q.weight q8_0 blk.0.attn_v.weight q8_0 -blk.1.ffn_down_shexp.weight q8_0 -blk.1.ffn_gate_shexp.weight q8_0 -blk.1.ffn_up_shexp.weight q8_0 +blk.0.ffn_down.weight q8_0 +blk.0.ffn_gate.weight q8_0 +blk.0.ffn_up.weight q8_0 blk.1.attn_k.weight q8_0 blk.1.attn_output.weight q8_0 blk.1.attn_q.weight q8_0 blk.1.attn_v.weight q8_0 -blk.2.ffn_down_shexp.weight q8_0 -blk.2.ffn_gate_shexp.weight q8_0 -blk.2.ffn_up_shexp.weight q8_0 +blk.1.ffn_down_shexp.weight q8_0 +blk.1.ffn_gate_shexp.weight q8_0 +blk.1.ffn_up_shexp.weight q8_0 blk.2.attn_k.weight q8_0 blk.2.attn_output.weight q8_0 blk.2.attn_q.weight q8_0 blk.2.attn_v.weight q8_0 -blk.3.ffn_down_shexp.weight q8_0 -blk.3.ffn_gate_shexp.weight q8_0 -blk.3.ffn_up_shexp.weight q8_0 +blk.2.ffn_down_shexp.weight q8_0 +blk.2.ffn_gate_shexp.weight q8_0 +blk.2.ffn_up_shexp.weight q8_0 blk.3.attn_k.weight q8_0 blk.3.attn_output.weight q8_0 blk.3.attn_q.weight q8_0 blk.3.attn_v.weight q8_0 -blk.4.ffn_down_shexp.weight q8_0 -blk.4.ffn_gate_shexp.weight q8_0 -blk.4.ffn_up_shexp.weight q8_0 +blk.3.ffn_down_shexp.weight q8_0 +blk.3.ffn_gate_shexp.weight q8_0 +blk.3.ffn_up_shexp.weight q8_0 blk.4.attn_k.weight q8_0 blk.4.attn_output.weight q8_0 blk.4.attn_q.weight q8_0 blk.4.attn_v.weight q8_0 -blk.5.ffn_down_shexp.weight q8_0 -blk.5.ffn_gate_shexp.weight q8_0 -blk.5.ffn_up_shexp.weight q8_0 +blk.4.ffn_down_shexp.weight q8_0 +blk.4.ffn_gate_shexp.weight q8_0 +blk.4.ffn_up_shexp.weight q8_0 blk.5.attn_k.weight q8_0 blk.5.attn_output.weight q8_0 blk.5.attn_q.weight q8_0 blk.5.attn_v.weight q8_0 -blk.6.ffn_down_shexp.weight q8_0 -blk.6.ffn_gate_shexp.weight q8_0 -blk.6.ffn_up_shexp.weight q8_0 +blk.5.ffn_down_shexp.weight q8_0 +blk.5.ffn_gate_shexp.weight q8_0 +blk.5.ffn_up_shexp.weight q8_0 blk.6.attn_k.weight q8_0 blk.6.attn_output.weight q8_0 blk.6.attn_q.weight q8_0 blk.6.attn_v.weight q8_0 -blk.7.ffn_down_shexp.weight q8_0 -blk.7.ffn_gate_shexp.weight q8_0 -blk.7.ffn_up_shexp.weight q8_0 +blk.6.ffn_down_shexp.weight q8_0 +blk.6.ffn_gate_shexp.weight q8_0 +blk.6.ffn_up_shexp.weight q8_0 blk.7.attn_k.weight q8_0 blk.7.attn_output.weight q8_0 blk.7.attn_q.weight q8_0 blk.7.attn_v.weight q8_0 -blk.8.ffn_down_shexp.weight q8_0 -blk.8.ffn_gate_shexp.weight q8_0 -blk.8.ffn_up_shexp.weight q8_0 +blk.7.ffn_down_shexp.weight q8_0 +blk.7.ffn_gate_shexp.weight q8_0 +blk.7.ffn_up_shexp.weight q8_0 blk.8.attn_k.weight q8_0 blk.8.attn_output.weight q8_0 blk.8.attn_q.weight q8_0 blk.8.attn_v.weight q8_0 -blk.9.ffn_down_shexp.weight q8_0 -blk.9.ffn_gate_shexp.weight q8_0 -blk.9.ffn_up_shexp.weight q8_0 +blk.8.ffn_down_shexp.weight q8_0 +blk.8.ffn_gate_shexp.weight q8_0 +blk.8.ffn_up_shexp.weight q8_0 blk.9.attn_k.weight q8_0 blk.9.attn_output.weight q8_0 blk.9.attn_q.weight q8_0 blk.9.attn_v.weight q8_0 -blk.10.ffn_down_shexp.weight q8_0 -blk.10.ffn_gate_shexp.weight q8_0 -blk.10.ffn_up_shexp.weight q8_0 +blk.9.ffn_down_shexp.weight q8_0 +blk.9.ffn_gate_shexp.weight q8_0 +blk.9.ffn_up_shexp.weight q8_0 blk.10.attn_k.weight q8_0 blk.10.attn_output.weight q8_0 blk.10.attn_q.weight q8_0 blk.10.attn_v.weight q8_0 -blk.11.ffn_down_shexp.weight q8_0 -blk.11.ffn_gate_shexp.weight q8_0 -blk.11.ffn_up_shexp.weight q8_0 +blk.10.ffn_down_shexp.weight q8_0 +blk.10.ffn_gate_shexp.weight q8_0 +blk.10.ffn_up_shexp.weight q8_0 blk.11.attn_k.weight q8_0 blk.11.attn_output.weight q8_0 blk.11.attn_q.weight q8_0 blk.11.attn_v.weight q8_0 -blk.12.ffn_down_shexp.weight q8_0 -blk.12.ffn_gate_shexp.weight q8_0 -blk.12.ffn_up_shexp.weight q8_0 +blk.11.ffn_down_shexp.weight q8_0 +blk.11.ffn_gate_shexp.weight q8_0 +blk.11.ffn_up_shexp.weight q8_0 blk.12.attn_k.weight q8_0 blk.12.attn_output.weight q8_0 blk.12.attn_q.weight q8_0 blk.12.attn_v.weight q8_0 -blk.13.ffn_down_shexp.weight q8_0 -blk.13.ffn_gate_shexp.weight q8_0 -blk.13.ffn_up_shexp.weight q8_0 +blk.12.ffn_down_shexp.weight q8_0 +blk.12.ffn_gate_shexp.weight q8_0 +blk.12.ffn_up_shexp.weight q8_0 blk.13.attn_k.weight q8_0 blk.13.attn_output.weight q8_0 blk.13.attn_q.weight q8_0 blk.13.attn_v.weight q8_0 -blk.14.ffn_down_shexp.weight q8_0 -blk.14.ffn_gate_shexp.weight q8_0 -blk.14.ffn_up_shexp.weight q8_0 +blk.13.ffn_down_shexp.weight q8_0 +blk.13.ffn_gate_shexp.weight q8_0 +blk.13.ffn_up_shexp.weight q8_0 blk.14.attn_k.weight q8_0 blk.14.attn_output.weight q8_0 blk.14.attn_q.weight q8_0 blk.14.attn_v.weight q8_0 -blk.15.ffn_down_shexp.weight q8_0 -blk.15.ffn_gate_shexp.weight q8_0 -blk.15.ffn_up_shexp.weight q8_0 +blk.14.ffn_down_shexp.weight q8_0 +blk.14.ffn_gate_shexp.weight q8_0 +blk.14.ffn_up_shexp.weight q8_0 blk.15.attn_k.weight q8_0 blk.15.attn_output.weight q8_0 blk.15.attn_q.weight q8_0 blk.15.attn_v.weight q8_0 -blk.16.ffn_down_shexp.weight q8_0 -blk.16.ffn_gate_shexp.weight q8_0 -blk.16.ffn_up_shexp.weight q8_0 +blk.15.ffn_down_shexp.weight q8_0 +blk.15.ffn_gate_shexp.weight q8_0 +blk.15.ffn_up_shexp.weight q8_0 blk.16.attn_k.weight q8_0 blk.16.attn_output.weight q8_0 blk.16.attn_q.weight q8_0 blk.16.attn_v.weight q8_0 -blk.17.ffn_down_shexp.weight q8_0 -blk.17.ffn_gate_shexp.weight q8_0 -blk.17.ffn_up_shexp.weight q8_0 +blk.16.ffn_down_shexp.weight q8_0 +blk.16.ffn_gate_shexp.weight q8_0 +blk.16.ffn_up_shexp.weight q8_0 blk.17.attn_k.weight q8_0 blk.17.attn_output.weight q8_0 blk.17.attn_q.weight q8_0 blk.17.attn_v.weight q8_0 -blk.18.ffn_down_shexp.weight q8_0 -blk.18.ffn_gate_shexp.weight q8_0 -blk.18.ffn_up_shexp.weight q8_0 +blk.17.ffn_down_shexp.weight q8_0 +blk.17.ffn_gate_shexp.weight q8_0 +blk.17.ffn_up_shexp.weight q8_0 blk.18.attn_k.weight q8_0 blk.18.attn_output.weight q8_0 blk.18.attn_q.weight q8_0 blk.18.attn_v.weight q8_0 -blk.19.ffn_down_shexp.weight q8_0 -blk.19.ffn_gate_shexp.weight q8_0 -blk.19.ffn_up_shexp.weight q8_0 +blk.18.ffn_down_shexp.weight q8_0 +blk.18.ffn_gate_shexp.weight q8_0 +blk.18.ffn_up_shexp.weight q8_0 blk.19.attn_k.weight q8_0 blk.19.attn_output.weight q8_0 blk.19.attn_q.weight q8_0 blk.19.attn_v.weight q8_0 -blk.20.ffn_down_shexp.weight q8_0 -blk.20.ffn_gate_shexp.weight q8_0 -blk.20.ffn_up_shexp.weight q8_0 +blk.19.ffn_down_shexp.weight q8_0 +blk.19.ffn_gate_shexp.weight q8_0 +blk.19.ffn_up_shexp.weight q8_0 blk.20.attn_k.weight q8_0 blk.20.attn_output.weight q8_0 blk.20.attn_q.weight q8_0 blk.20.attn_v.weight q8_0 -blk.21.ffn_down_shexp.weight q8_0 -blk.21.ffn_gate_shexp.weight q8_0 -blk.21.ffn_up_shexp.weight q8_0 +blk.20.ffn_down_shexp.weight q8_0 +blk.20.ffn_gate_shexp.weight q8_0 +blk.20.ffn_up_shexp.weight q8_0 blk.21.attn_k.weight q8_0 blk.21.attn_output.weight q8_0 blk.21.attn_q.weight q8_0 blk.21.attn_v.weight q8_0 -blk.22.ffn_down_shexp.weight q8_0 -blk.22.ffn_gate_shexp.weight q8_0 -blk.22.ffn_up_shexp.weight q8_0 +blk.21.ffn_down_shexp.weight q8_0 +blk.21.ffn_gate_shexp.weight q8_0 +blk.21.ffn_up_shexp.weight q8_0 blk.22.attn_k.weight q8_0 blk.22.attn_output.weight q8_0 blk.22.attn_q.weight q8_0 blk.22.attn_v.weight q8_0 -blk.23.ffn_down_shexp.weight q8_0 -blk.23.ffn_gate_shexp.weight q8_0 -blk.23.ffn_up_shexp.weight q8_0 +blk.22.ffn_down_shexp.weight q8_0 +blk.22.ffn_gate_shexp.weight q8_0 +blk.22.ffn_up_shexp.weight q8_0 blk.23.attn_k.weight q8_0 blk.23.attn_output.weight q8_0 blk.23.attn_q.weight q8_0 blk.23.attn_v.weight q8_0 -blk.24.ffn_down_shexp.weight q8_0 -blk.24.ffn_gate_shexp.weight q8_0 -blk.24.ffn_up_shexp.weight q8_0 +blk.23.ffn_down_shexp.weight q8_0 +blk.23.ffn_gate_shexp.weight q8_0 +blk.23.ffn_up_shexp.weight q8_0 blk.24.attn_k.weight q8_0 blk.24.attn_output.weight q8_0 blk.24.attn_q.weight q8_0 blk.24.attn_v.weight q8_0 -blk.25.ffn_down_shexp.weight q8_0 -blk.25.ffn_gate_shexp.weight q8_0 -blk.25.ffn_up_shexp.weight q8_0 +blk.24.ffn_down_shexp.weight q8_0 +blk.24.ffn_gate_shexp.weight q8_0 +blk.24.ffn_up_shexp.weight q8_0 blk.25.attn_k.weight q8_0 blk.25.attn_output.weight q8_0 blk.25.attn_q.weight q8_0 blk.25.attn_v.weight q8_0 -blk.26.ffn_down_shexp.weight q8_0 -blk.26.ffn_gate_shexp.weight q8_0 -blk.26.ffn_up_shexp.weight q8_0 +blk.25.ffn_down_shexp.weight q8_0 +blk.25.ffn_gate_shexp.weight q8_0 +blk.25.ffn_up_shexp.weight q8_0 blk.26.attn_k.weight q8_0 blk.26.attn_output.weight q8_0 blk.26.attn_q.weight q8_0 blk.26.attn_v.weight q8_0 -blk.27.ffn_down_shexp.weight q8_0 -blk.27.ffn_gate_shexp.weight q8_0 -blk.27.ffn_up_shexp.weight q8_0 +blk.26.ffn_down_shexp.weight q8_0 +blk.26.ffn_gate_shexp.weight q8_0 +blk.26.ffn_up_shexp.weight q8_0 blk.27.attn_k.weight q8_0 blk.27.attn_output.weight q8_0 blk.27.attn_q.weight q8_0 blk.27.attn_v.weight q8_0 -blk.28.ffn_down_shexp.weight q8_0 -blk.28.ffn_gate_shexp.weight q8_0 -blk.28.ffn_up_shexp.weight q8_0 +blk.27.ffn_down_shexp.weight q8_0 +blk.27.ffn_gate_shexp.weight q8_0 +blk.27.ffn_up_shexp.weight q8_0 blk.28.attn_k.weight q8_0 blk.28.attn_output.weight q8_0 blk.28.attn_q.weight q8_0 blk.28.attn_v.weight q8_0 -blk.29.ffn_down_shexp.weight q8_0 -blk.29.ffn_gate_shexp.weight q8_0 -blk.29.ffn_up_shexp.weight q8_0 +blk.28.ffn_down_shexp.weight q8_0 +blk.28.ffn_gate_shexp.weight q8_0 +blk.28.ffn_up_shexp.weight q8_0 blk.29.attn_k.weight q8_0 blk.29.attn_output.weight q8_0 blk.29.attn_q.weight q8_0 blk.29.attn_v.weight q8_0 -blk.30.ffn_down_shexp.weight q8_0 -blk.30.ffn_gate_shexp.weight q8_0 -blk.30.ffn_up_shexp.weight q8_0 +blk.29.ffn_down_shexp.weight q8_0 +blk.29.ffn_gate_shexp.weight q8_0 +blk.29.ffn_up_shexp.weight q8_0 blk.30.attn_k.weight q8_0 blk.30.attn_output.weight q8_0 blk.30.attn_q.weight q8_0 blk.30.attn_v.weight q8_0 -blk.31.ffn_down_shexp.weight q8_0 -blk.31.ffn_gate_shexp.weight q8_0 -blk.31.ffn_up_shexp.weight q8_0 +blk.30.ffn_down_shexp.weight q8_0 +blk.30.ffn_gate_shexp.weight q8_0 +blk.30.ffn_up_shexp.weight q8_0 blk.31.attn_k.weight q8_0 blk.31.attn_output.weight q8_0 blk.31.attn_q.weight q8_0 blk.31.attn_v.weight q8_0 -blk.32.ffn_down_shexp.weight q8_0 -blk.32.ffn_gate_shexp.weight q8_0 -blk.32.ffn_up_shexp.weight q8_0 +blk.31.ffn_down_shexp.weight q8_0 +blk.31.ffn_gate_shexp.weight q8_0 +blk.31.ffn_up_shexp.weight q8_0 blk.32.attn_k.weight q8_0 blk.32.attn_output.weight q8_0 blk.32.attn_q.weight q8_0 blk.32.attn_v.weight q8_0 -blk.33.ffn_down_shexp.weight q8_0 -blk.33.ffn_gate_shexp.weight q8_0 -blk.33.ffn_up_shexp.weight q8_0 +blk.32.ffn_down_shexp.weight q8_0 +blk.32.ffn_gate_shexp.weight q8_0 +blk.32.ffn_up_shexp.weight q8_0 blk.33.attn_k.weight q8_0 blk.33.attn_output.weight q8_0 blk.33.attn_q.weight q8_0 blk.33.attn_v.weight q8_0 -blk.34.ffn_gate_shexp.weight q8_0 -blk.34.attn_output.weight q8_0 -blk.34.ffn_down_shexp.weight q8_0 -blk.34.ffn_up_shexp.weight q8_0 +blk.33.ffn_down_shexp.weight q8_0 +blk.33.ffn_gate_shexp.weight q8_0 +blk.33.ffn_up_shexp.weight q8_0 blk.34.attn_k.weight q8_0 +blk.34.attn_output.weight q8_0 blk.34.attn_q.weight q8_0 blk.34.attn_v.weight q8_0 -blk.35.ffn_down_shexp.weight q8_0 -blk.35.ffn_gate_shexp.weight q8_0 -blk.35.ffn_up_shexp.weight q8_0 +blk.34.ffn_down_shexp.weight q8_0 +blk.34.ffn_gate_shexp.weight q8_0 +blk.34.ffn_up_shexp.weight q8_0 blk.35.attn_k.weight q8_0 blk.35.attn_output.weight q8_0 blk.35.attn_q.weight q8_0 blk.35.attn_v.weight q8_0 -blk.36.ffn_down_shexp.weight q8_0 -blk.36.ffn_gate_shexp.weight q8_0 -blk.36.ffn_up_shexp.weight q8_0 +blk.35.ffn_down_shexp.weight q8_0 +blk.35.ffn_gate_shexp.weight q8_0 +blk.35.ffn_up_shexp.weight q8_0 blk.36.attn_k.weight q8_0 blk.36.attn_output.weight q8_0 blk.36.attn_q.weight q8_0 blk.36.attn_v.weight q8_0 -blk.37.ffn_down_shexp.weight q8_0 -blk.37.ffn_gate_shexp.weight q8_0 -blk.37.ffn_up_shexp.weight q8_0 +blk.36.ffn_down_shexp.weight q8_0 +blk.36.ffn_gate_shexp.weight q8_0 +blk.36.ffn_up_shexp.weight q8_0 blk.37.attn_k.weight q8_0 blk.37.attn_output.weight q8_0 blk.37.attn_q.weight q8_0 blk.37.attn_v.weight q8_0 -blk.38.ffn_down_shexp.weight q8_0 -blk.38.ffn_gate_shexp.weight q8_0 -blk.38.ffn_up_shexp.weight q8_0 +blk.37.ffn_down_shexp.weight q8_0 +blk.37.ffn_gate_shexp.weight q8_0 +blk.37.ffn_up_shexp.weight q8_0 blk.38.attn_k.weight q8_0 blk.38.attn_output.weight q8_0 blk.38.attn_q.weight q8_0 blk.38.attn_v.weight q8_0 -blk.39.ffn_down_shexp.weight q8_0 -blk.39.ffn_gate_shexp.weight q8_0 -blk.39.ffn_up_shexp.weight q8_0 +blk.38.ffn_down_shexp.weight q8_0 +blk.38.ffn_gate_shexp.weight q8_0 +blk.38.ffn_up_shexp.weight q8_0 blk.39.attn_k.weight q8_0 blk.39.attn_output.weight q8_0 blk.39.attn_q.weight q8_0 blk.39.attn_v.weight q8_0 -blk.40.ffn_down_shexp.weight q8_0 -blk.40.ffn_gate_shexp.weight q8_0 -blk.40.ffn_up_shexp.weight q8_0 +blk.39.ffn_down_shexp.weight q8_0 +blk.39.ffn_gate_shexp.weight q8_0 +blk.39.ffn_up_shexp.weight q8_0 blk.40.attn_k.weight q8_0 blk.40.attn_output.weight q8_0 blk.40.attn_q.weight q8_0 blk.40.attn_v.weight q8_0 -blk.41.ffn_down_shexp.weight q8_0 -blk.41.ffn_gate_shexp.weight q8_0 -blk.41.ffn_up_shexp.weight q8_0 +blk.40.ffn_down_shexp.weight q8_0 +blk.40.ffn_gate_shexp.weight q8_0 +blk.40.ffn_up_shexp.weight q8_0 blk.41.attn_k.weight q8_0 blk.41.attn_output.weight q8_0 blk.41.attn_q.weight q8_0 blk.41.attn_v.weight q8_0 +blk.41.ffn_down_shexp.weight q8_0 +blk.41.ffn_gate_shexp.weight q8_0 +blk.41.ffn_up_shexp.weight q8_0 +blk.42.attn_k.weight q8_0 +blk.42.attn_output.weight q8_0 +blk.42.attn_q.weight q8_0 +blk.42.attn_v.weight q8_0 blk.42.ffn_down_shexp.weight q8_0 blk.42.ffn_gate_shexp.weight q8_0 blk.42.ffn_up_shexp.weight q8_0 -blk.42.attn_k.weight q8_0 -blk.42.attn_q.weight q8_0 -blk.42.attn_v.weight q8_0 -blk.42.attn_output.weight q8_0 -blk.43.ffn_down_shexp.weight q8_0 -blk.43.ffn_gate_shexp.weight q8_0 -blk.43.ffn_up_shexp.weight q8_0 blk.43.attn_k.weight q8_0 blk.43.attn_output.weight q8_0 blk.43.attn_q.weight q8_0 blk.43.attn_v.weight q8_0 -blk.44.ffn_down_shexp.weight q8_0 -blk.44.ffn_gate_shexp.weight q8_0 -blk.44.ffn_up_shexp.weight q8_0 +blk.43.ffn_down_shexp.weight q8_0 +blk.43.ffn_gate_shexp.weight q8_0 +blk.43.ffn_up_shexp.weight q8_0 blk.44.attn_k.weight q8_0 blk.44.attn_output.weight q8_0 blk.44.attn_q.weight q8_0 blk.44.attn_v.weight q8_0 -output.weight q8_0 -blk.45.ffn_down_shexp.weight q8_0 -blk.45.ffn_gate_shexp.weight q8_0 -blk.45.ffn_up_shexp.weight q8_0 +blk.44.ffn_down_shexp.weight q8_0 +blk.44.ffn_gate_shexp.weight q8_0 +blk.44.ffn_up_shexp.weight q8_0 blk.45.attn_k.weight q8_0 blk.45.attn_output.weight q8_0 blk.45.attn_q.weight q8_0 blk.45.attn_v.weight q8_0 +blk.45.ffn_down_shexp.weight q8_0 +blk.45.ffn_gate_shexp.weight q8_0 +blk.45.ffn_up_shexp.weight q8_0 diff --git a/tests/snapshots/meta-llama-3.1-70b-instruct.schema b/tests/snapshots/meta-llama-3.1-70b-instruct.schema index 95e0fdf148..36e4d90ac6 100644 --- a/tests/snapshots/meta-llama-3.1-70b-instruct.schema +++ b/tests/snapshots/meta-llama-3.1-70b-instruct.schema @@ -20,249 +20,250 @@ output.weight q6_K output.weight q6_K [Q2_K] q2_K -blk.0.ffn_down.weight q3_K +output.weight q6_K blk.0.attn_output.weight q3_K blk.0.attn_v.weight q5_K +blk.0.ffn_down.weight q3_K blk.1.attn_output.weight q3_K blk.1.attn_v.weight q5_K blk.1.ffn_down.weight q3_K -blk.2.ffn_down.weight q3_K blk.2.attn_output.weight q3_K blk.2.attn_v.weight q5_K -blk.3.ffn_down.weight q3_K +blk.2.ffn_down.weight q3_K blk.3.attn_output.weight q3_K blk.3.attn_v.weight q5_K +blk.3.ffn_down.weight q3_K blk.4.attn_output.weight q3_K blk.4.attn_v.weight q5_K blk.4.ffn_down.weight q3_K -blk.5.ffn_down.weight q3_K blk.5.attn_output.weight q3_K blk.5.attn_v.weight q5_K -blk.6.ffn_down.weight q3_K +blk.5.ffn_down.weight q3_K blk.6.attn_output.weight q3_K blk.6.attn_v.weight q5_K +blk.6.ffn_down.weight q3_K +blk.7.attn_output.weight q3_K blk.7.attn_v.weight q5_K blk.7.ffn_down.weight q3_K -blk.7.attn_output.weight q3_K -blk.8.ffn_down.weight q3_K blk.8.attn_output.weight q3_K blk.8.attn_v.weight q5_K -blk.9.ffn_down.weight q3_K +blk.8.ffn_down.weight q3_K blk.9.attn_output.weight q3_K blk.9.attn_v.weight q5_K -blk.10.ffn_down.weight q3_K +blk.9.ffn_down.weight q3_K blk.10.attn_output.weight q3_K blk.10.attn_v.weight q5_K -blk.11.ffn_down.weight q3_K +blk.10.ffn_down.weight q3_K blk.11.attn_output.weight q3_K blk.11.attn_v.weight q5_K +blk.11.ffn_down.weight q3_K blk.12.attn_output.weight q3_K blk.12.attn_v.weight q5_K blk.12.ffn_down.weight q3_K -blk.13.ffn_down.weight q3_K blk.13.attn_output.weight q3_K blk.13.attn_v.weight q5_K -blk.14.ffn_down.weight q3_K +blk.13.ffn_down.weight q3_K blk.14.attn_output.weight q3_K blk.14.attn_v.weight q5_K +blk.14.ffn_down.weight q3_K blk.15.attn_output.weight q3_K blk.15.attn_v.weight q5_K blk.15.ffn_down.weight q3_K -blk.16.ffn_down.weight q3_K blk.16.attn_output.weight q3_K blk.16.attn_v.weight q5_K -blk.17.ffn_down.weight q3_K +blk.16.ffn_down.weight q3_K blk.17.attn_output.weight q3_K blk.17.attn_v.weight q5_K +blk.17.ffn_down.weight q3_K blk.18.attn_output.weight q3_K blk.18.attn_v.weight q5_K blk.18.ffn_down.weight q3_K -blk.19.ffn_down.weight q3_K blk.19.attn_output.weight q3_K blk.19.attn_v.weight q5_K -blk.20.ffn_down.weight q3_K +blk.19.ffn_down.weight q3_K blk.20.attn_output.weight q3_K blk.20.attn_v.weight q5_K +blk.20.ffn_down.weight q3_K +blk.21.attn_output.weight q3_K blk.21.attn_v.weight q5_K blk.21.ffn_down.weight q3_K -blk.21.attn_output.weight q3_K -blk.22.ffn_down.weight q3_K blk.22.attn_output.weight q3_K blk.22.attn_v.weight q5_K -blk.23.ffn_down.weight q3_K +blk.22.ffn_down.weight q3_K blk.23.attn_output.weight q3_K blk.23.attn_v.weight q5_K -blk.24.ffn_down.weight q3_K +blk.23.ffn_down.weight q3_K blk.24.attn_output.weight q3_K blk.24.attn_v.weight q5_K -blk.25.ffn_down.weight q3_K +blk.24.ffn_down.weight q3_K blk.25.attn_output.weight q3_K blk.25.attn_v.weight q5_K +blk.25.ffn_down.weight q3_K blk.26.attn_output.weight q3_K blk.26.attn_v.weight q5_K blk.26.ffn_down.weight q3_K -blk.27.ffn_down.weight q3_K blk.27.attn_output.weight q3_K blk.27.attn_v.weight q5_K -blk.28.ffn_down.weight q3_K +blk.27.ffn_down.weight q3_K blk.28.attn_output.weight q3_K blk.28.attn_v.weight q5_K +blk.28.ffn_down.weight q3_K blk.29.attn_output.weight q3_K blk.29.attn_v.weight q5_K blk.29.ffn_down.weight q3_K -blk.30.ffn_down.weight q3_K blk.30.attn_output.weight q3_K blk.30.attn_v.weight q5_K -blk.31.ffn_down.weight q3_K +blk.30.ffn_down.weight q3_K blk.31.attn_output.weight q3_K blk.31.attn_v.weight q5_K +blk.31.ffn_down.weight q3_K blk.32.attn_output.weight q3_K blk.32.attn_v.weight q5_K blk.32.ffn_down.weight q3_K -blk.33.ffn_down.weight q3_K blk.33.attn_output.weight q3_K blk.33.attn_v.weight q5_K -blk.34.ffn_down.weight q3_K +blk.33.ffn_down.weight q3_K blk.34.attn_output.weight q3_K blk.34.attn_v.weight q5_K +blk.34.ffn_down.weight q3_K +blk.35.attn_output.weight q3_K blk.35.attn_v.weight q5_K blk.35.ffn_down.weight q3_K -blk.35.attn_output.weight q3_K -blk.36.ffn_down.weight q3_K blk.36.attn_output.weight q3_K blk.36.attn_v.weight q5_K -blk.37.ffn_down.weight q3_K +blk.36.ffn_down.weight q3_K blk.37.attn_output.weight q3_K blk.37.attn_v.weight q5_K -blk.38.ffn_down.weight q3_K +blk.37.ffn_down.weight q3_K blk.38.attn_output.weight q3_K blk.38.attn_v.weight q5_K -blk.39.ffn_down.weight q3_K +blk.38.ffn_down.weight q3_K blk.39.attn_output.weight q3_K blk.39.attn_v.weight q5_K +blk.39.ffn_down.weight q3_K blk.40.attn_output.weight q3_K blk.40.attn_v.weight q5_K blk.40.ffn_down.weight q3_K -blk.41.ffn_down.weight q3_K blk.41.attn_output.weight q3_K blk.41.attn_v.weight q5_K -blk.42.ffn_down.weight q3_K +blk.41.ffn_down.weight q3_K blk.42.attn_output.weight q3_K blk.42.attn_v.weight q5_K +blk.42.ffn_down.weight q3_K blk.43.attn_output.weight q3_K blk.43.attn_v.weight q5_K blk.43.ffn_down.weight q3_K -blk.44.ffn_down.weight q3_K blk.44.attn_output.weight q3_K blk.44.attn_v.weight q5_K -blk.45.ffn_down.weight q3_K +blk.44.ffn_down.weight q3_K blk.45.attn_output.weight q3_K blk.45.attn_v.weight q5_K +blk.45.ffn_down.weight q3_K blk.46.attn_output.weight q3_K blk.46.attn_v.weight q5_K blk.46.ffn_down.weight q3_K -blk.47.ffn_down.weight q3_K blk.47.attn_output.weight q3_K blk.47.attn_v.weight q5_K -blk.48.ffn_down.weight q3_K +blk.47.ffn_down.weight q3_K blk.48.attn_output.weight q3_K blk.48.attn_v.weight q5_K +blk.48.ffn_down.weight q3_K +blk.49.attn_output.weight q3_K blk.49.attn_v.weight q5_K blk.49.ffn_down.weight q3_K -blk.49.attn_output.weight q3_K -blk.50.ffn_down.weight q3_K blk.50.attn_output.weight q3_K blk.50.attn_v.weight q5_K -blk.51.ffn_down.weight q3_K +blk.50.ffn_down.weight q3_K blk.51.attn_output.weight q3_K blk.51.attn_v.weight q5_K -blk.52.ffn_down.weight q3_K +blk.51.ffn_down.weight q3_K blk.52.attn_output.weight q3_K blk.52.attn_v.weight q5_K -blk.53.ffn_down.weight q3_K +blk.52.ffn_down.weight q3_K blk.53.attn_output.weight q3_K blk.53.attn_v.weight q5_K +blk.53.ffn_down.weight q3_K blk.54.attn_output.weight q3_K blk.54.attn_v.weight q5_K blk.54.ffn_down.weight q3_K -blk.55.ffn_down.weight q3_K blk.55.attn_output.weight q3_K blk.55.attn_v.weight q5_K -blk.56.ffn_down.weight q3_K +blk.55.ffn_down.weight q3_K blk.56.attn_output.weight q3_K blk.56.attn_v.weight q5_K +blk.56.ffn_down.weight q3_K blk.57.attn_output.weight q3_K blk.57.attn_v.weight q5_K blk.57.ffn_down.weight q3_K -blk.58.ffn_down.weight q3_K blk.58.attn_output.weight q3_K blk.58.attn_v.weight q5_K -blk.59.ffn_down.weight q3_K +blk.58.ffn_down.weight q3_K blk.59.attn_output.weight q3_K blk.59.attn_v.weight q5_K +blk.59.ffn_down.weight q3_K blk.60.attn_output.weight q3_K blk.60.attn_v.weight q5_K blk.60.ffn_down.weight q3_K -blk.61.ffn_down.weight q3_K blk.61.attn_output.weight q3_K blk.61.attn_v.weight q5_K -blk.62.ffn_down.weight q3_K +blk.61.ffn_down.weight q3_K blk.62.attn_output.weight q3_K blk.62.attn_v.weight q5_K +blk.62.ffn_down.weight q3_K +blk.63.attn_output.weight q3_K blk.63.attn_v.weight q5_K blk.63.ffn_down.weight q3_K -blk.63.attn_output.weight q3_K -blk.64.ffn_down.weight q3_K blk.64.attn_output.weight q3_K blk.64.attn_v.weight q5_K -blk.65.ffn_down.weight q3_K +blk.64.ffn_down.weight q3_K blk.65.attn_output.weight q3_K blk.65.attn_v.weight q5_K -blk.66.ffn_down.weight q3_K +blk.65.ffn_down.weight q3_K blk.66.attn_output.weight q3_K blk.66.attn_v.weight q5_K -blk.67.ffn_down.weight q3_K +blk.66.ffn_down.weight q3_K blk.67.attn_output.weight q3_K blk.67.attn_v.weight q5_K +blk.67.ffn_down.weight q3_K blk.68.attn_output.weight q3_K blk.68.attn_v.weight q5_K blk.68.ffn_down.weight q3_K -blk.69.ffn_down.weight q3_K blk.69.attn_output.weight q3_K blk.69.attn_v.weight q5_K -blk.70.ffn_down.weight q3_K +blk.69.ffn_down.weight q3_K blk.70.attn_output.weight q3_K blk.70.attn_v.weight q5_K +blk.70.ffn_down.weight q3_K blk.71.attn_output.weight q3_K blk.71.attn_v.weight q5_K blk.71.ffn_down.weight q3_K -blk.72.ffn_down.weight q3_K blk.72.attn_output.weight q3_K blk.72.attn_v.weight q5_K -blk.73.ffn_down.weight q3_K +blk.72.ffn_down.weight q3_K blk.73.attn_output.weight q3_K blk.73.attn_v.weight q5_K +blk.73.ffn_down.weight q3_K blk.74.attn_output.weight q3_K blk.74.attn_v.weight q5_K blk.74.ffn_down.weight q3_K -blk.75.ffn_down.weight q3_K blk.75.attn_output.weight q3_K blk.75.attn_v.weight q5_K -blk.76.ffn_down.weight q3_K +blk.75.ffn_down.weight q3_K blk.76.attn_output.weight q3_K blk.76.attn_v.weight q5_K +blk.76.ffn_down.weight q3_K +blk.77.attn_output.weight q3_K blk.77.attn_v.weight q5_K blk.77.ffn_down.weight q3_K -blk.77.attn_output.weight q3_K -blk.78.ffn_down.weight q3_K blk.78.attn_output.weight q3_K blk.78.attn_v.weight q5_K -blk.79.ffn_down.weight q3_K +blk.78.ffn_down.weight q3_K blk.79.attn_output.weight q3_K blk.79.attn_v.weight q5_K -output.weight q6_K +blk.79.ffn_down.weight q3_K [Q3_K_S] q3_K +output.weight q6_K blk.0.attn_v.weight q5_K blk.1.attn_v.weight q5_K blk.2.attn_v.weight q5_K @@ -343,515 +344,515 @@ blk.76.attn_v.weight q5_K blk.77.attn_v.weight q5_K blk.78.attn_v.weight q5_K blk.79.attn_v.weight q5_K -output.weight q6_K [Q3_K_M] q3_K -blk.0.ffn_down.weight q5_K +output.weight q6_K blk.0.attn_output.weight q4_K blk.0.attn_v.weight q5_K +blk.0.ffn_down.weight q5_K blk.1.attn_output.weight q4_K blk.1.attn_v.weight q5_K blk.1.ffn_down.weight q5_K -blk.2.ffn_down.weight q5_K blk.2.attn_output.weight q4_K blk.2.attn_v.weight q5_K -blk.3.ffn_down.weight q5_K +blk.2.ffn_down.weight q5_K blk.3.attn_output.weight q4_K blk.3.attn_v.weight q5_K +blk.3.ffn_down.weight q5_K blk.4.attn_output.weight q4_K blk.4.attn_v.weight q5_K blk.4.ffn_down.weight q5_K -blk.5.ffn_down.weight q4_K blk.5.attn_output.weight q4_K blk.5.attn_v.weight q5_K -blk.6.ffn_down.weight q4_K +blk.5.ffn_down.weight q4_K blk.6.attn_output.weight q4_K blk.6.attn_v.weight q5_K +blk.6.ffn_down.weight q4_K +blk.7.attn_output.weight q4_K blk.7.attn_v.weight q5_K blk.7.ffn_down.weight q4_K -blk.7.attn_output.weight q4_K -blk.8.ffn_down.weight q4_K blk.8.attn_output.weight q4_K blk.8.attn_v.weight q5_K -blk.9.ffn_down.weight q4_K +blk.8.ffn_down.weight q4_K blk.9.attn_output.weight q4_K blk.9.attn_v.weight q5_K -blk.10.ffn_down.weight q4_K +blk.9.ffn_down.weight q4_K blk.10.attn_output.weight q4_K blk.10.attn_v.weight q5_K -blk.11.ffn_down.weight q4_K +blk.10.ffn_down.weight q4_K blk.11.attn_output.weight q4_K blk.11.attn_v.weight q5_K +blk.11.ffn_down.weight q4_K blk.12.attn_output.weight q4_K blk.12.attn_v.weight q5_K blk.12.ffn_down.weight q4_K -blk.13.ffn_down.weight q4_K blk.13.attn_output.weight q4_K blk.13.attn_v.weight q5_K -blk.14.ffn_down.weight q4_K +blk.13.ffn_down.weight q4_K blk.14.attn_output.weight q4_K blk.14.attn_v.weight q5_K +blk.14.ffn_down.weight q4_K blk.15.attn_output.weight q4_K blk.15.attn_v.weight q5_K blk.15.ffn_down.weight q4_K -blk.16.ffn_down.weight q4_K blk.16.attn_output.weight q4_K blk.16.attn_v.weight q5_K -blk.17.ffn_down.weight q4_K +blk.16.ffn_down.weight q4_K blk.17.attn_output.weight q4_K blk.17.attn_v.weight q5_K +blk.17.ffn_down.weight q4_K blk.18.attn_output.weight q4_K blk.18.attn_v.weight q5_K blk.18.ffn_down.weight q4_K -blk.19.ffn_down.weight q4_K blk.19.attn_output.weight q4_K blk.19.attn_v.weight q5_K -blk.20.ffn_down.weight q4_K +blk.19.ffn_down.weight q4_K blk.20.attn_output.weight q4_K blk.20.attn_v.weight q5_K +blk.20.ffn_down.weight q4_K +blk.21.attn_output.weight q4_K blk.21.attn_v.weight q5_K blk.21.ffn_down.weight q4_K -blk.21.attn_output.weight q4_K -blk.22.ffn_down.weight q4_K blk.22.attn_output.weight q4_K blk.22.attn_v.weight q5_K -blk.23.ffn_down.weight q4_K +blk.22.ffn_down.weight q4_K blk.23.attn_output.weight q4_K blk.23.attn_v.weight q5_K -blk.24.ffn_down.weight q4_K +blk.23.ffn_down.weight q4_K blk.24.attn_output.weight q4_K blk.24.attn_v.weight q5_K -blk.25.ffn_down.weight q4_K +blk.24.ffn_down.weight q4_K blk.25.attn_output.weight q4_K blk.25.attn_v.weight q5_K +blk.25.ffn_down.weight q4_K blk.26.attn_output.weight q4_K blk.26.attn_v.weight q5_K blk.26.ffn_down.weight q4_K -blk.27.ffn_down.weight q4_K blk.27.attn_output.weight q4_K blk.27.attn_v.weight q5_K -blk.28.ffn_down.weight q4_K +blk.27.ffn_down.weight q4_K blk.28.attn_output.weight q4_K blk.28.attn_v.weight q5_K +blk.28.ffn_down.weight q4_K blk.29.attn_output.weight q4_K blk.29.attn_v.weight q5_K blk.29.ffn_down.weight q4_K -blk.30.ffn_down.weight q4_K blk.30.attn_output.weight q4_K blk.30.attn_v.weight q5_K -blk.31.ffn_down.weight q4_K +blk.30.ffn_down.weight q4_K blk.31.attn_output.weight q4_K blk.31.attn_v.weight q5_K +blk.31.ffn_down.weight q4_K blk.32.attn_output.weight q4_K blk.32.attn_v.weight q5_K blk.32.ffn_down.weight q4_K -blk.33.ffn_down.weight q4_K blk.33.attn_output.weight q4_K blk.33.attn_v.weight q5_K -blk.34.ffn_down.weight q4_K +blk.33.ffn_down.weight q4_K blk.34.attn_output.weight q4_K blk.34.attn_v.weight q5_K +blk.34.ffn_down.weight q4_K +blk.35.attn_output.weight q4_K blk.35.attn_v.weight q5_K blk.35.ffn_down.weight q4_K -blk.35.attn_output.weight q4_K -blk.36.ffn_down.weight q4_K blk.36.attn_output.weight q4_K blk.36.attn_v.weight q5_K -blk.37.ffn_down.weight q4_K +blk.36.ffn_down.weight q4_K blk.37.attn_output.weight q4_K blk.37.attn_v.weight q5_K -blk.38.ffn_down.weight q4_K +blk.37.ffn_down.weight q4_K blk.38.attn_output.weight q4_K blk.38.attn_v.weight q5_K -blk.39.ffn_down.weight q4_K +blk.38.ffn_down.weight q4_K blk.39.attn_output.weight q4_K blk.39.attn_v.weight q5_K +blk.39.ffn_down.weight q4_K blk.40.attn_output.weight q4_K blk.40.attn_v.weight q5_K blk.40.ffn_down.weight q4_K -blk.41.ffn_down.weight q4_K blk.41.attn_output.weight q4_K blk.41.attn_v.weight q5_K -blk.42.ffn_down.weight q4_K +blk.41.ffn_down.weight q4_K blk.42.attn_output.weight q4_K blk.42.attn_v.weight q5_K +blk.42.ffn_down.weight q4_K blk.43.attn_output.weight q4_K blk.43.attn_v.weight q5_K blk.43.ffn_down.weight q4_K -blk.44.ffn_down.weight q4_K blk.44.attn_output.weight q4_K blk.44.attn_v.weight q5_K -blk.45.ffn_down.weight q4_K +blk.44.ffn_down.weight q4_K blk.45.attn_output.weight q4_K blk.45.attn_v.weight q5_K +blk.45.ffn_down.weight q4_K blk.46.attn_output.weight q4_K blk.46.attn_v.weight q5_K blk.46.ffn_down.weight q4_K -blk.47.ffn_down.weight q4_K blk.47.attn_output.weight q4_K blk.47.attn_v.weight q5_K -blk.48.ffn_down.weight q4_K +blk.47.ffn_down.weight q4_K blk.48.attn_output.weight q4_K blk.48.attn_v.weight q5_K +blk.48.ffn_down.weight q4_K +blk.49.attn_output.weight q4_K blk.49.attn_v.weight q5_K blk.49.ffn_down.weight q4_K -blk.49.attn_output.weight q4_K -blk.50.ffn_down.weight q4_K blk.50.attn_output.weight q4_K blk.50.attn_v.weight q5_K -blk.51.ffn_down.weight q4_K +blk.50.ffn_down.weight q4_K blk.51.attn_output.weight q4_K blk.51.attn_v.weight q5_K -blk.52.ffn_down.weight q4_K +blk.51.ffn_down.weight q4_K blk.52.attn_output.weight q4_K blk.52.attn_v.weight q5_K -blk.53.ffn_down.weight q4_K +blk.52.ffn_down.weight q4_K blk.53.attn_output.weight q4_K blk.53.attn_v.weight q5_K +blk.53.ffn_down.weight q4_K blk.54.attn_output.weight q4_K blk.54.attn_v.weight q5_K blk.54.ffn_down.weight q4_K -blk.55.ffn_down.weight q4_K blk.55.attn_output.weight q4_K blk.55.attn_v.weight q5_K -blk.56.ffn_down.weight q4_K +blk.55.ffn_down.weight q4_K blk.56.attn_output.weight q4_K blk.56.attn_v.weight q5_K +blk.56.ffn_down.weight q4_K blk.57.attn_output.weight q4_K blk.57.attn_v.weight q5_K blk.57.ffn_down.weight q4_K -blk.58.ffn_down.weight q4_K blk.58.attn_output.weight q4_K blk.58.attn_v.weight q5_K -blk.59.ffn_down.weight q4_K +blk.58.ffn_down.weight q4_K blk.59.attn_output.weight q4_K blk.59.attn_v.weight q5_K +blk.59.ffn_down.weight q4_K blk.60.attn_output.weight q4_K blk.60.attn_v.weight q5_K blk.60.ffn_down.weight q4_K -blk.61.ffn_down.weight q4_K blk.61.attn_output.weight q4_K blk.61.attn_v.weight q5_K -blk.62.ffn_down.weight q4_K +blk.61.ffn_down.weight q4_K blk.62.attn_output.weight q4_K blk.62.attn_v.weight q5_K +blk.62.ffn_down.weight q4_K +blk.63.attn_output.weight q4_K blk.63.attn_v.weight q5_K blk.63.ffn_down.weight q4_K -blk.63.attn_output.weight q4_K -blk.64.ffn_down.weight q4_K blk.64.attn_output.weight q4_K blk.64.attn_v.weight q5_K -blk.65.ffn_down.weight q4_K +blk.64.ffn_down.weight q4_K blk.65.attn_output.weight q4_K blk.65.attn_v.weight q5_K -blk.66.ffn_down.weight q4_K +blk.65.ffn_down.weight q4_K blk.66.attn_output.weight q4_K blk.66.attn_v.weight q5_K -blk.67.ffn_down.weight q4_K +blk.66.ffn_down.weight q4_K blk.67.attn_output.weight q4_K blk.67.attn_v.weight q5_K +blk.67.ffn_down.weight q4_K blk.68.attn_output.weight q4_K blk.68.attn_v.weight q5_K blk.68.ffn_down.weight q4_K -blk.69.ffn_down.weight q4_K blk.69.attn_output.weight q4_K blk.69.attn_v.weight q5_K -blk.70.ffn_down.weight q4_K +blk.69.ffn_down.weight q4_K blk.70.attn_output.weight q4_K blk.70.attn_v.weight q5_K +blk.70.ffn_down.weight q4_K blk.71.attn_output.weight q4_K blk.71.attn_v.weight q5_K blk.71.ffn_down.weight q4_K -blk.72.ffn_down.weight q4_K blk.72.attn_output.weight q4_K blk.72.attn_v.weight q5_K -blk.73.ffn_down.weight q4_K +blk.72.ffn_down.weight q4_K blk.73.attn_output.weight q4_K blk.73.attn_v.weight q5_K +blk.73.ffn_down.weight q4_K blk.74.attn_output.weight q4_K blk.74.attn_v.weight q5_K blk.74.ffn_down.weight q4_K -blk.75.ffn_down.weight q4_K blk.75.attn_output.weight q4_K blk.75.attn_v.weight q5_K -blk.76.ffn_down.weight q4_K +blk.75.ffn_down.weight q4_K blk.76.attn_output.weight q4_K blk.76.attn_v.weight q5_K +blk.76.ffn_down.weight q4_K +blk.77.attn_output.weight q4_K blk.77.attn_v.weight q5_K blk.77.ffn_down.weight q4_K -blk.77.attn_output.weight q4_K -blk.78.ffn_down.weight q4_K blk.78.attn_output.weight q4_K blk.78.attn_v.weight q5_K -blk.79.ffn_down.weight q4_K +blk.78.ffn_down.weight q4_K blk.79.attn_output.weight q4_K blk.79.attn_v.weight q5_K -output.weight q6_K +blk.79.ffn_down.weight q4_K [Q3_K_L] q3_K -blk.0.ffn_down.weight q5_K +output.weight q6_K blk.0.attn_output.weight q5_K blk.0.attn_v.weight q5_K +blk.0.ffn_down.weight q5_K blk.1.attn_output.weight q5_K blk.1.attn_v.weight q5_K blk.1.ffn_down.weight q5_K -blk.2.ffn_down.weight q5_K blk.2.attn_output.weight q5_K blk.2.attn_v.weight q5_K -blk.3.ffn_down.weight q5_K +blk.2.ffn_down.weight q5_K blk.3.attn_output.weight q5_K blk.3.attn_v.weight q5_K +blk.3.ffn_down.weight q5_K blk.4.attn_output.weight q5_K blk.4.attn_v.weight q5_K blk.4.ffn_down.weight q5_K -blk.5.ffn_down.weight q5_K blk.5.attn_output.weight q5_K blk.5.attn_v.weight q5_K -blk.6.ffn_down.weight q5_K +blk.5.ffn_down.weight q5_K blk.6.attn_output.weight q5_K blk.6.attn_v.weight q5_K +blk.6.ffn_down.weight q5_K +blk.7.attn_output.weight q5_K blk.7.attn_v.weight q5_K blk.7.ffn_down.weight q5_K -blk.7.attn_output.weight q5_K -blk.8.ffn_down.weight q5_K blk.8.attn_output.weight q5_K blk.8.attn_v.weight q5_K -blk.9.ffn_down.weight q5_K +blk.8.ffn_down.weight q5_K blk.9.attn_output.weight q5_K blk.9.attn_v.weight q5_K -blk.10.ffn_down.weight q5_K +blk.9.ffn_down.weight q5_K blk.10.attn_output.weight q5_K blk.10.attn_v.weight q5_K -blk.11.ffn_down.weight q5_K +blk.10.ffn_down.weight q5_K blk.11.attn_output.weight q5_K blk.11.attn_v.weight q5_K +blk.11.ffn_down.weight q5_K blk.12.attn_output.weight q5_K blk.12.attn_v.weight q5_K blk.12.ffn_down.weight q5_K -blk.13.ffn_down.weight q5_K blk.13.attn_output.weight q5_K blk.13.attn_v.weight q5_K -blk.14.ffn_down.weight q5_K +blk.13.ffn_down.weight q5_K blk.14.attn_output.weight q5_K blk.14.attn_v.weight q5_K +blk.14.ffn_down.weight q5_K blk.15.attn_output.weight q5_K blk.15.attn_v.weight q5_K blk.15.ffn_down.weight q5_K -blk.16.ffn_down.weight q5_K blk.16.attn_output.weight q5_K blk.16.attn_v.weight q5_K -blk.17.ffn_down.weight q5_K +blk.16.ffn_down.weight q5_K blk.17.attn_output.weight q5_K blk.17.attn_v.weight q5_K +blk.17.ffn_down.weight q5_K blk.18.attn_output.weight q5_K blk.18.attn_v.weight q5_K blk.18.ffn_down.weight q5_K -blk.19.ffn_down.weight q5_K blk.19.attn_output.weight q5_K blk.19.attn_v.weight q5_K -blk.20.ffn_down.weight q5_K +blk.19.ffn_down.weight q5_K blk.20.attn_output.weight q5_K blk.20.attn_v.weight q5_K +blk.20.ffn_down.weight q5_K +blk.21.attn_output.weight q5_K blk.21.attn_v.weight q5_K blk.21.ffn_down.weight q5_K -blk.21.attn_output.weight q5_K -blk.22.ffn_down.weight q5_K blk.22.attn_output.weight q5_K blk.22.attn_v.weight q5_K -blk.23.ffn_down.weight q5_K +blk.22.ffn_down.weight q5_K blk.23.attn_output.weight q5_K blk.23.attn_v.weight q5_K -blk.24.ffn_down.weight q5_K +blk.23.ffn_down.weight q5_K blk.24.attn_output.weight q5_K blk.24.attn_v.weight q5_K -blk.25.ffn_down.weight q5_K +blk.24.ffn_down.weight q5_K blk.25.attn_output.weight q5_K blk.25.attn_v.weight q5_K +blk.25.ffn_down.weight q5_K blk.26.attn_output.weight q5_K blk.26.attn_v.weight q5_K blk.26.ffn_down.weight q5_K -blk.27.ffn_down.weight q5_K blk.27.attn_output.weight q5_K blk.27.attn_v.weight q5_K -blk.28.ffn_down.weight q5_K +blk.27.ffn_down.weight q5_K blk.28.attn_output.weight q5_K blk.28.attn_v.weight q5_K +blk.28.ffn_down.weight q5_K blk.29.attn_output.weight q5_K blk.29.attn_v.weight q5_K blk.29.ffn_down.weight q5_K -blk.30.ffn_down.weight q5_K blk.30.attn_output.weight q5_K blk.30.attn_v.weight q5_K -blk.31.ffn_down.weight q5_K +blk.30.ffn_down.weight q5_K blk.31.attn_output.weight q5_K blk.31.attn_v.weight q5_K +blk.31.ffn_down.weight q5_K blk.32.attn_output.weight q5_K blk.32.attn_v.weight q5_K blk.32.ffn_down.weight q5_K -blk.33.ffn_down.weight q5_K blk.33.attn_output.weight q5_K blk.33.attn_v.weight q5_K -blk.34.ffn_down.weight q5_K +blk.33.ffn_down.weight q5_K blk.34.attn_output.weight q5_K blk.34.attn_v.weight q5_K +blk.34.ffn_down.weight q5_K +blk.35.attn_output.weight q5_K blk.35.attn_v.weight q5_K blk.35.ffn_down.weight q5_K -blk.35.attn_output.weight q5_K -blk.36.ffn_down.weight q5_K blk.36.attn_output.weight q5_K blk.36.attn_v.weight q5_K -blk.37.ffn_down.weight q5_K +blk.36.ffn_down.weight q5_K blk.37.attn_output.weight q5_K blk.37.attn_v.weight q5_K -blk.38.ffn_down.weight q5_K +blk.37.ffn_down.weight q5_K blk.38.attn_output.weight q5_K blk.38.attn_v.weight q5_K -blk.39.ffn_down.weight q5_K +blk.38.ffn_down.weight q5_K blk.39.attn_output.weight q5_K blk.39.attn_v.weight q5_K +blk.39.ffn_down.weight q5_K blk.40.attn_output.weight q5_K blk.40.attn_v.weight q5_K blk.40.ffn_down.weight q5_K -blk.41.ffn_down.weight q5_K blk.41.attn_output.weight q5_K blk.41.attn_v.weight q5_K -blk.42.ffn_down.weight q5_K +blk.41.ffn_down.weight q5_K blk.42.attn_output.weight q5_K blk.42.attn_v.weight q5_K +blk.42.ffn_down.weight q5_K blk.43.attn_output.weight q5_K blk.43.attn_v.weight q5_K blk.43.ffn_down.weight q5_K -blk.44.ffn_down.weight q5_K blk.44.attn_output.weight q5_K blk.44.attn_v.weight q5_K -blk.45.ffn_down.weight q5_K +blk.44.ffn_down.weight q5_K blk.45.attn_output.weight q5_K blk.45.attn_v.weight q5_K +blk.45.ffn_down.weight q5_K blk.46.attn_output.weight q5_K blk.46.attn_v.weight q5_K blk.46.ffn_down.weight q5_K -blk.47.ffn_down.weight q5_K blk.47.attn_output.weight q5_K blk.47.attn_v.weight q5_K -blk.48.ffn_down.weight q5_K +blk.47.ffn_down.weight q5_K blk.48.attn_output.weight q5_K blk.48.attn_v.weight q5_K +blk.48.ffn_down.weight q5_K +blk.49.attn_output.weight q5_K blk.49.attn_v.weight q5_K blk.49.ffn_down.weight q5_K -blk.49.attn_output.weight q5_K -blk.50.ffn_down.weight q5_K blk.50.attn_output.weight q5_K blk.50.attn_v.weight q5_K -blk.51.ffn_down.weight q5_K +blk.50.ffn_down.weight q5_K blk.51.attn_output.weight q5_K blk.51.attn_v.weight q5_K -blk.52.ffn_down.weight q5_K +blk.51.ffn_down.weight q5_K blk.52.attn_output.weight q5_K blk.52.attn_v.weight q5_K -blk.53.ffn_down.weight q5_K +blk.52.ffn_down.weight q5_K blk.53.attn_output.weight q5_K blk.53.attn_v.weight q5_K +blk.53.ffn_down.weight q5_K blk.54.attn_output.weight q5_K blk.54.attn_v.weight q5_K blk.54.ffn_down.weight q5_K -blk.55.ffn_down.weight q5_K blk.55.attn_output.weight q5_K blk.55.attn_v.weight q5_K -blk.56.ffn_down.weight q5_K +blk.55.ffn_down.weight q5_K blk.56.attn_output.weight q5_K blk.56.attn_v.weight q5_K +blk.56.ffn_down.weight q5_K blk.57.attn_output.weight q5_K blk.57.attn_v.weight q5_K blk.57.ffn_down.weight q5_K -blk.58.ffn_down.weight q5_K blk.58.attn_output.weight q5_K blk.58.attn_v.weight q5_K -blk.59.ffn_down.weight q5_K +blk.58.ffn_down.weight q5_K blk.59.attn_output.weight q5_K blk.59.attn_v.weight q5_K +blk.59.ffn_down.weight q5_K blk.60.attn_output.weight q5_K blk.60.attn_v.weight q5_K blk.60.ffn_down.weight q5_K -blk.61.ffn_down.weight q5_K blk.61.attn_output.weight q5_K blk.61.attn_v.weight q5_K -blk.62.ffn_down.weight q5_K +blk.61.ffn_down.weight q5_K blk.62.attn_output.weight q5_K blk.62.attn_v.weight q5_K +blk.62.ffn_down.weight q5_K +blk.63.attn_output.weight q5_K blk.63.attn_v.weight q5_K blk.63.ffn_down.weight q5_K -blk.63.attn_output.weight q5_K -blk.64.ffn_down.weight q5_K blk.64.attn_output.weight q5_K blk.64.attn_v.weight q5_K -blk.65.ffn_down.weight q5_K +blk.64.ffn_down.weight q5_K blk.65.attn_output.weight q5_K blk.65.attn_v.weight q5_K -blk.66.ffn_down.weight q5_K +blk.65.ffn_down.weight q5_K blk.66.attn_output.weight q5_K blk.66.attn_v.weight q5_K -blk.67.ffn_down.weight q5_K +blk.66.ffn_down.weight q5_K blk.67.attn_output.weight q5_K blk.67.attn_v.weight q5_K +blk.67.ffn_down.weight q5_K blk.68.attn_output.weight q5_K blk.68.attn_v.weight q5_K blk.68.ffn_down.weight q5_K -blk.69.ffn_down.weight q5_K blk.69.attn_output.weight q5_K blk.69.attn_v.weight q5_K -blk.70.ffn_down.weight q5_K +blk.69.ffn_down.weight q5_K blk.70.attn_output.weight q5_K blk.70.attn_v.weight q5_K +blk.70.ffn_down.weight q5_K blk.71.attn_output.weight q5_K blk.71.attn_v.weight q5_K blk.71.ffn_down.weight q5_K -blk.72.ffn_down.weight q5_K blk.72.attn_output.weight q5_K blk.72.attn_v.weight q5_K -blk.73.ffn_down.weight q5_K +blk.72.ffn_down.weight q5_K blk.73.attn_output.weight q5_K blk.73.attn_v.weight q5_K +blk.73.ffn_down.weight q5_K blk.74.attn_output.weight q5_K blk.74.attn_v.weight q5_K blk.74.ffn_down.weight q5_K -blk.75.ffn_down.weight q5_K blk.75.attn_output.weight q5_K blk.75.attn_v.weight q5_K -blk.76.ffn_down.weight q5_K +blk.75.ffn_down.weight q5_K blk.76.attn_output.weight q5_K blk.76.attn_v.weight q5_K +blk.76.ffn_down.weight q5_K +blk.77.attn_output.weight q5_K blk.77.attn_v.weight q5_K blk.77.ffn_down.weight q5_K -blk.77.attn_output.weight q5_K -blk.78.ffn_down.weight q5_K blk.78.attn_output.weight q5_K blk.78.attn_v.weight q5_K -blk.79.ffn_down.weight q5_K +blk.78.ffn_down.weight q5_K blk.79.attn_output.weight q5_K blk.79.attn_v.weight q5_K -output.weight q6_K +blk.79.ffn_down.weight q5_K [Q4_K_S] q4_K -blk.0.ffn_down.weight q5_K +output.weight q6_K blk.0.attn_v.weight q5_K +blk.0.ffn_down.weight q5_K blk.1.attn_v.weight q5_K blk.1.ffn_down.weight q5_K -blk.2.ffn_down.weight q5_K blk.2.attn_v.weight q5_K -blk.3.ffn_down.weight q5_K +blk.2.ffn_down.weight q5_K blk.3.attn_v.weight q5_K +blk.3.ffn_down.weight q5_K blk.4.attn_v.weight q5_K blk.4.ffn_down.weight q5_K -blk.5.ffn_down.weight q5_K blk.5.attn_v.weight q5_K -blk.6.ffn_down.weight q5_K +blk.5.ffn_down.weight q5_K blk.6.attn_v.weight q5_K +blk.6.ffn_down.weight q5_K blk.7.attn_v.weight q5_K blk.7.ffn_down.weight q5_K -blk.8.ffn_down.weight q5_K blk.8.attn_v.weight q5_K -blk.9.ffn_down.weight q5_K +blk.8.ffn_down.weight q5_K blk.9.attn_v.weight q5_K +blk.9.ffn_down.weight q5_K blk.10.attn_v.weight q5_K blk.11.attn_v.weight q5_K blk.12.attn_v.weight q5_K @@ -922,29 +923,29 @@ blk.76.attn_v.weight q5_K blk.77.attn_v.weight q5_K blk.78.attn_v.weight q5_K blk.79.attn_v.weight q5_K -output.weight q6_K [Q4_K_M] q4_K -blk.0.ffn_down.weight q6_K +output.weight q6_K blk.0.attn_v.weight q6_K +blk.0.ffn_down.weight q6_K blk.1.attn_v.weight q6_K blk.1.ffn_down.weight q6_K -blk.2.ffn_down.weight q6_K blk.2.attn_v.weight q6_K -blk.3.ffn_down.weight q6_K +blk.2.ffn_down.weight q6_K blk.3.attn_v.weight q6_K +blk.3.ffn_down.weight q6_K blk.4.attn_v.weight q6_K blk.4.ffn_down.weight q6_K -blk.5.ffn_down.weight q6_K blk.5.attn_v.weight q6_K -blk.6.ffn_down.weight q6_K +blk.5.ffn_down.weight q6_K blk.6.attn_v.weight q6_K +blk.6.ffn_down.weight q6_K blk.7.attn_v.weight q6_K blk.7.ffn_down.weight q6_K -blk.8.ffn_down.weight q6_K blk.8.attn_v.weight q6_K -blk.9.ffn_down.weight q6_K +blk.8.ffn_down.weight q6_K blk.9.attn_v.weight q6_K +blk.9.ffn_down.weight q6_K blk.10.attn_v.weight q5_K blk.11.attn_v.weight q5_K blk.12.attn_v.weight q6_K @@ -963,44 +964,44 @@ blk.21.attn_v.weight q6_K blk.21.ffn_down.weight q6_K blk.22.attn_v.weight q5_K blk.23.attn_v.weight q5_K -blk.24.ffn_down.weight q6_K blk.24.attn_v.weight q6_K +blk.24.ffn_down.weight q6_K blk.25.attn_v.weight q5_K blk.26.attn_v.weight q5_K -blk.27.ffn_down.weight q6_K blk.27.attn_v.weight q6_K +blk.27.ffn_down.weight q6_K blk.28.attn_v.weight q5_K blk.29.attn_v.weight q5_K -blk.30.ffn_down.weight q6_K blk.30.attn_v.weight q6_K +blk.30.ffn_down.weight q6_K blk.31.attn_v.weight q5_K blk.32.attn_v.weight q5_K -blk.33.ffn_down.weight q6_K blk.33.attn_v.weight q6_K +blk.33.ffn_down.weight q6_K blk.34.attn_v.weight q5_K blk.35.attn_v.weight q5_K -blk.36.ffn_down.weight q6_K blk.36.attn_v.weight q6_K +blk.36.ffn_down.weight q6_K blk.37.attn_v.weight q5_K blk.38.attn_v.weight q5_K -blk.39.ffn_down.weight q6_K blk.39.attn_v.weight q6_K +blk.39.ffn_down.weight q6_K blk.40.attn_v.weight q5_K blk.41.attn_v.weight q5_K -blk.42.ffn_down.weight q6_K blk.42.attn_v.weight q6_K +blk.42.ffn_down.weight q6_K blk.43.attn_v.weight q5_K blk.44.attn_v.weight q5_K -blk.45.ffn_down.weight q6_K blk.45.attn_v.weight q6_K +blk.45.ffn_down.weight q6_K blk.46.attn_v.weight q5_K blk.47.attn_v.weight q5_K -blk.48.ffn_down.weight q6_K blk.48.attn_v.weight q6_K +blk.48.ffn_down.weight q6_K blk.49.attn_v.weight q5_K blk.50.attn_v.weight q5_K -blk.51.ffn_down.weight q6_K blk.51.attn_v.weight q6_K +blk.51.ffn_down.weight q6_K blk.52.attn_v.weight q5_K blk.53.attn_v.weight q5_K blk.54.attn_v.weight q6_K @@ -1019,58 +1020,58 @@ blk.63.attn_v.weight q6_K blk.63.ffn_down.weight q6_K blk.64.attn_v.weight q5_K blk.65.attn_v.weight q5_K -blk.66.ffn_down.weight q6_K blk.66.attn_v.weight q6_K +blk.66.ffn_down.weight q6_K blk.67.attn_v.weight q5_K blk.68.attn_v.weight q5_K -blk.69.ffn_down.weight q6_K blk.69.attn_v.weight q6_K -blk.70.ffn_down.weight q6_K +blk.69.ffn_down.weight q6_K blk.70.attn_v.weight q6_K +blk.70.ffn_down.weight q6_K blk.71.attn_v.weight q6_K blk.71.ffn_down.weight q6_K -blk.72.ffn_down.weight q6_K blk.72.attn_v.weight q6_K -blk.73.ffn_down.weight q6_K +blk.72.ffn_down.weight q6_K blk.73.attn_v.weight q6_K +blk.73.ffn_down.weight q6_K blk.74.attn_v.weight q6_K blk.74.ffn_down.weight q6_K -blk.75.ffn_down.weight q6_K blk.75.attn_v.weight q6_K -blk.76.ffn_down.weight q6_K +blk.75.ffn_down.weight q6_K blk.76.attn_v.weight q6_K +blk.76.ffn_down.weight q6_K blk.77.attn_v.weight q6_K blk.77.ffn_down.weight q6_K -blk.78.ffn_down.weight q6_K blk.78.attn_v.weight q6_K -blk.79.ffn_down.weight q6_K +blk.78.ffn_down.weight q6_K blk.79.attn_v.weight q6_K -output.weight q6_K +blk.79.ffn_down.weight q6_K [Q5_K_S] q5_K output.weight q6_K [Q5_K_M] q5_K -blk.0.ffn_down.weight q6_K +output.weight q6_K blk.0.attn_v.weight q6_K +blk.0.ffn_down.weight q6_K blk.1.attn_v.weight q6_K blk.1.ffn_down.weight q6_K -blk.2.ffn_down.weight q6_K blk.2.attn_v.weight q6_K -blk.3.ffn_down.weight q6_K +blk.2.ffn_down.weight q6_K blk.3.attn_v.weight q6_K +blk.3.ffn_down.weight q6_K blk.4.attn_v.weight q6_K blk.4.ffn_down.weight q6_K -blk.5.ffn_down.weight q6_K blk.5.attn_v.weight q6_K -blk.6.ffn_down.weight q6_K +blk.5.ffn_down.weight q6_K blk.6.attn_v.weight q6_K +blk.6.ffn_down.weight q6_K blk.7.attn_v.weight q6_K blk.7.ffn_down.weight q6_K -blk.8.ffn_down.weight q6_K blk.8.attn_v.weight q6_K -blk.9.ffn_down.weight q6_K +blk.8.ffn_down.weight q6_K blk.9.attn_v.weight q6_K +blk.9.ffn_down.weight q6_K blk.12.attn_v.weight q6_K blk.12.ffn_down.weight q6_K blk.15.attn_v.weight q6_K @@ -1079,26 +1080,26 @@ blk.18.attn_v.weight q6_K blk.18.ffn_down.weight q6_K blk.21.attn_v.weight q6_K blk.21.ffn_down.weight q6_K -blk.24.ffn_down.weight q6_K blk.24.attn_v.weight q6_K -blk.27.ffn_down.weight q6_K +blk.24.ffn_down.weight q6_K blk.27.attn_v.weight q6_K -blk.30.ffn_down.weight q6_K +blk.27.ffn_down.weight q6_K blk.30.attn_v.weight q6_K -blk.33.ffn_down.weight q6_K +blk.30.ffn_down.weight q6_K blk.33.attn_v.weight q6_K -blk.36.ffn_down.weight q6_K +blk.33.ffn_down.weight q6_K blk.36.attn_v.weight q6_K -blk.39.ffn_down.weight q6_K +blk.36.ffn_down.weight q6_K blk.39.attn_v.weight q6_K -blk.42.ffn_down.weight q6_K +blk.39.ffn_down.weight q6_K blk.42.attn_v.weight q6_K -blk.45.ffn_down.weight q6_K +blk.42.ffn_down.weight q6_K blk.45.attn_v.weight q6_K -blk.48.ffn_down.weight q6_K +blk.45.ffn_down.weight q6_K blk.48.attn_v.weight q6_K -blk.51.ffn_down.weight q6_K +blk.48.ffn_down.weight q6_K blk.51.attn_v.weight q6_K +blk.51.ffn_down.weight q6_K blk.54.attn_v.weight q6_K blk.54.ffn_down.weight q6_K blk.57.attn_v.weight q6_K @@ -1107,56 +1108,56 @@ blk.60.attn_v.weight q6_K blk.60.ffn_down.weight q6_K blk.63.attn_v.weight q6_K blk.63.ffn_down.weight q6_K -blk.66.ffn_down.weight q6_K blk.66.attn_v.weight q6_K -blk.69.ffn_down.weight q6_K +blk.66.ffn_down.weight q6_K blk.69.attn_v.weight q6_K -blk.70.ffn_down.weight q6_K +blk.69.ffn_down.weight q6_K blk.70.attn_v.weight q6_K +blk.70.ffn_down.weight q6_K blk.71.attn_v.weight q6_K blk.71.ffn_down.weight q6_K -blk.72.ffn_down.weight q6_K blk.72.attn_v.weight q6_K -blk.73.ffn_down.weight q6_K +blk.72.ffn_down.weight q6_K blk.73.attn_v.weight q6_K +blk.73.ffn_down.weight q6_K blk.74.attn_v.weight q6_K blk.74.ffn_down.weight q6_K -blk.75.ffn_down.weight q6_K blk.75.attn_v.weight q6_K -blk.76.ffn_down.weight q6_K +blk.75.ffn_down.weight q6_K blk.76.attn_v.weight q6_K +blk.76.ffn_down.weight q6_K blk.77.attn_v.weight q6_K blk.77.ffn_down.weight q6_K -blk.78.ffn_down.weight q6_K blk.78.attn_v.weight q6_K -blk.79.ffn_down.weight q6_K +blk.78.ffn_down.weight q6_K blk.79.attn_v.weight q6_K -output.weight q6_K +blk.79.ffn_down.weight q6_K [Q6_K] q6_K [IQ2_XXS] iq2_xxs +output.weight q5_K token_embd.weight q2_K -blk.0.ffn_down.weight q2_K blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q2_K blk.1.attn_v.weight q4_K blk.1.ffn_down.weight q2_K -blk.2.ffn_down.weight q2_K blk.2.attn_v.weight q4_K -blk.3.ffn_down.weight q2_K +blk.2.ffn_down.weight q2_K blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight q2_K blk.4.attn_v.weight q4_K blk.4.ffn_down.weight q2_K -blk.5.ffn_down.weight q2_K blk.5.attn_v.weight q4_K -blk.6.ffn_down.weight q2_K +blk.5.ffn_down.weight q2_K blk.6.attn_v.weight q4_K +blk.6.ffn_down.weight q2_K blk.7.attn_v.weight q4_K blk.7.ffn_down.weight q2_K -blk.8.ffn_down.weight q2_K blk.8.attn_v.weight q4_K -blk.9.ffn_down.weight q2_K +blk.8.ffn_down.weight q2_K blk.9.attn_v.weight q4_K +blk.9.ffn_down.weight q2_K blk.10.attn_v.weight q4_K blk.11.attn_v.weight q4_K blk.12.attn_v.weight q4_K @@ -1227,30 +1228,30 @@ blk.76.attn_v.weight q4_K blk.77.attn_v.weight q4_K blk.78.attn_v.weight q4_K blk.79.attn_v.weight q4_K -output.weight q5_K [IQ2_XS] iq2_xs +output.weight q5_K token_embd.weight q2_K -blk.0.ffn_down.weight q2_K blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q2_K blk.1.attn_v.weight q4_K blk.1.ffn_down.weight q2_K -blk.2.ffn_down.weight q2_K blk.2.attn_v.weight q4_K -blk.3.ffn_down.weight q2_K +blk.2.ffn_down.weight q2_K blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight q2_K blk.4.attn_v.weight q4_K blk.4.ffn_down.weight q2_K -blk.5.ffn_down.weight q2_K blk.5.attn_v.weight q4_K -blk.6.ffn_down.weight q2_K +blk.5.ffn_down.weight q2_K blk.6.attn_v.weight q4_K +blk.6.ffn_down.weight q2_K blk.7.attn_v.weight q4_K blk.7.ffn_down.weight q2_K -blk.8.ffn_down.weight q2_K blk.8.attn_v.weight q4_K -blk.9.ffn_down.weight q2_K +blk.8.ffn_down.weight q2_K blk.9.attn_v.weight q4_K +blk.9.ffn_down.weight q2_K blk.10.attn_v.weight q4_K blk.11.attn_v.weight q4_K blk.12.attn_v.weight q4_K @@ -1321,29 +1322,29 @@ blk.76.attn_v.weight q4_K blk.77.attn_v.weight q4_K blk.78.attn_v.weight q4_K blk.79.attn_v.weight q4_K -output.weight q5_K [Q2_K_S] q2_K -blk.0.ffn_down.weight q4_K +output.weight q6_K blk.0.attn_v.weight q5_K +blk.0.ffn_down.weight q4_K blk.1.attn_v.weight q5_K blk.1.ffn_down.weight q4_K -blk.2.ffn_down.weight q4_K blk.2.attn_v.weight q5_K -blk.3.ffn_down.weight q4_K +blk.2.ffn_down.weight q4_K blk.3.attn_v.weight q5_K +blk.3.ffn_down.weight q4_K blk.4.attn_v.weight q5_K blk.4.ffn_down.weight q4_K -blk.5.ffn_down.weight q4_K blk.5.attn_v.weight q5_K -blk.6.ffn_down.weight q4_K +blk.5.ffn_down.weight q4_K blk.6.attn_v.weight q5_K +blk.6.ffn_down.weight q4_K blk.7.attn_v.weight q5_K blk.7.ffn_down.weight q4_K -blk.8.ffn_down.weight q4_K blk.8.attn_v.weight q5_K -blk.9.ffn_down.weight q4_K +blk.8.ffn_down.weight q4_K blk.9.attn_v.weight q5_K +blk.9.ffn_down.weight q4_K blk.10.attn_v.weight q5_K blk.11.attn_v.weight q5_K blk.12.attn_v.weight q5_K @@ -1414,9 +1415,9 @@ blk.76.attn_v.weight q5_K blk.77.attn_v.weight q5_K blk.78.attn_v.weight q5_K blk.79.attn_v.weight q5_K -output.weight q6_K [IQ3_XS] iq3_s +output.weight q6_K blk.0.attn_k.weight iq3_xxs blk.0.attn_q.weight iq3_xxs blk.0.attn_v.weight q5_K @@ -1447,306 +1448,306 @@ blk.8.attn_v.weight q5_K blk.9.attn_k.weight iq3_xxs blk.9.attn_q.weight iq3_xxs blk.9.attn_v.weight q5_K -blk.10.ffn_gate.weight iq3_xxs -blk.10.ffn_up.weight iq3_xxs blk.10.attn_k.weight iq3_xxs blk.10.attn_q.weight iq3_xxs blk.10.attn_v.weight q5_K -blk.11.ffn_gate.weight iq3_xxs -blk.11.ffn_up.weight iq3_xxs +blk.10.ffn_gate.weight iq3_xxs +blk.10.ffn_up.weight iq3_xxs blk.11.attn_k.weight iq3_xxs blk.11.attn_q.weight iq3_xxs blk.11.attn_v.weight q5_K -blk.12.ffn_gate.weight iq3_xxs -blk.12.ffn_up.weight iq3_xxs +blk.11.ffn_gate.weight iq3_xxs +blk.11.ffn_up.weight iq3_xxs blk.12.attn_k.weight iq3_xxs blk.12.attn_q.weight iq3_xxs blk.12.attn_v.weight q5_K -blk.13.ffn_gate.weight iq3_xxs -blk.13.ffn_up.weight iq3_xxs +blk.12.ffn_gate.weight iq3_xxs +blk.12.ffn_up.weight iq3_xxs blk.13.attn_k.weight iq3_xxs blk.13.attn_q.weight iq3_xxs blk.13.attn_v.weight q5_K -blk.14.ffn_gate.weight iq3_xxs -blk.14.ffn_up.weight iq3_xxs +blk.13.ffn_gate.weight iq3_xxs +blk.13.ffn_up.weight iq3_xxs blk.14.attn_k.weight iq3_xxs blk.14.attn_q.weight iq3_xxs blk.14.attn_v.weight q5_K -blk.15.ffn_gate.weight iq3_xxs +blk.14.ffn_gate.weight iq3_xxs +blk.14.ffn_up.weight iq3_xxs blk.15.attn_k.weight iq3_xxs blk.15.attn_q.weight iq3_xxs blk.15.attn_v.weight q5_K +blk.15.ffn_gate.weight iq3_xxs blk.15.ffn_up.weight iq3_xxs -blk.16.ffn_gate.weight iq3_xxs -blk.16.ffn_up.weight iq3_xxs blk.16.attn_k.weight iq3_xxs blk.16.attn_q.weight iq3_xxs blk.16.attn_v.weight q5_K -blk.17.ffn_gate.weight iq3_xxs -blk.17.ffn_up.weight iq3_xxs +blk.16.ffn_gate.weight iq3_xxs +blk.16.ffn_up.weight iq3_xxs blk.17.attn_k.weight iq3_xxs blk.17.attn_q.weight iq3_xxs blk.17.attn_v.weight q5_K +blk.17.ffn_gate.weight iq3_xxs +blk.17.ffn_up.weight iq3_xxs blk.18.attn_k.weight iq3_xxs blk.18.attn_q.weight iq3_xxs blk.18.attn_v.weight q5_K blk.18.ffn_gate.weight iq3_xxs blk.18.ffn_up.weight iq3_xxs -blk.19.ffn_gate.weight iq3_xxs -blk.19.ffn_up.weight iq3_xxs blk.19.attn_k.weight iq3_xxs blk.19.attn_q.weight iq3_xxs blk.19.attn_v.weight q5_K -blk.20.ffn_gate.weight iq3_xxs -blk.20.ffn_up.weight iq3_xxs +blk.19.ffn_gate.weight iq3_xxs +blk.19.ffn_up.weight iq3_xxs blk.20.attn_k.weight iq3_xxs blk.20.attn_q.weight iq3_xxs blk.20.attn_v.weight q5_K +blk.20.ffn_gate.weight iq3_xxs +blk.20.ffn_up.weight iq3_xxs blk.21.attn_k.weight iq3_xxs blk.21.attn_q.weight iq3_xxs blk.21.attn_v.weight q5_K blk.21.ffn_gate.weight iq3_xxs blk.21.ffn_up.weight iq3_xxs -blk.22.ffn_gate.weight iq3_xxs -blk.22.ffn_up.weight iq3_xxs blk.22.attn_k.weight iq3_xxs blk.22.attn_q.weight iq3_xxs blk.22.attn_v.weight q5_K -blk.23.ffn_gate.weight iq3_xxs -blk.23.ffn_up.weight iq3_xxs +blk.22.ffn_gate.weight iq3_xxs +blk.22.ffn_up.weight iq3_xxs blk.23.attn_k.weight iq3_xxs blk.23.attn_q.weight iq3_xxs blk.23.attn_v.weight q5_K -blk.24.ffn_gate.weight iq3_xxs -blk.24.ffn_up.weight iq3_xxs +blk.23.ffn_gate.weight iq3_xxs +blk.23.ffn_up.weight iq3_xxs blk.24.attn_k.weight iq3_xxs blk.24.attn_q.weight iq3_xxs blk.24.attn_v.weight q5_K -blk.25.ffn_gate.weight iq3_xxs -blk.25.ffn_up.weight iq3_xxs +blk.24.ffn_gate.weight iq3_xxs +blk.24.ffn_up.weight iq3_xxs blk.25.attn_k.weight iq3_xxs blk.25.attn_q.weight iq3_xxs blk.25.attn_v.weight q5_K -blk.26.ffn_gate.weight iq3_xxs -blk.26.ffn_up.weight iq3_xxs +blk.25.ffn_gate.weight iq3_xxs +blk.25.ffn_up.weight iq3_xxs blk.26.attn_k.weight iq3_xxs blk.26.attn_q.weight iq3_xxs blk.26.attn_v.weight q5_K -blk.27.ffn_gate.weight iq3_xxs -blk.27.ffn_up.weight iq3_xxs +blk.26.ffn_gate.weight iq3_xxs +blk.26.ffn_up.weight iq3_xxs blk.27.attn_k.weight iq3_xxs blk.27.attn_q.weight iq3_xxs blk.27.attn_v.weight q5_K -blk.28.ffn_gate.weight iq3_xxs -blk.28.ffn_up.weight iq3_xxs +blk.27.ffn_gate.weight iq3_xxs +blk.27.ffn_up.weight iq3_xxs blk.28.attn_k.weight iq3_xxs blk.28.attn_q.weight iq3_xxs blk.28.attn_v.weight q5_K -blk.29.ffn_gate.weight iq3_xxs +blk.28.ffn_gate.weight iq3_xxs +blk.28.ffn_up.weight iq3_xxs blk.29.attn_k.weight iq3_xxs blk.29.attn_q.weight iq3_xxs blk.29.attn_v.weight q5_K +blk.29.ffn_gate.weight iq3_xxs blk.29.ffn_up.weight iq3_xxs -blk.30.ffn_gate.weight iq3_xxs -blk.30.ffn_up.weight iq3_xxs blk.30.attn_k.weight iq3_xxs blk.30.attn_q.weight iq3_xxs blk.30.attn_v.weight q5_K -blk.31.ffn_gate.weight iq3_xxs -blk.31.ffn_up.weight iq3_xxs +blk.30.ffn_gate.weight iq3_xxs +blk.30.ffn_up.weight iq3_xxs blk.31.attn_k.weight iq3_xxs blk.31.attn_q.weight iq3_xxs blk.31.attn_v.weight q5_K +blk.31.ffn_gate.weight iq3_xxs +blk.31.ffn_up.weight iq3_xxs blk.32.attn_k.weight iq3_xxs blk.32.attn_q.weight iq3_xxs blk.32.attn_v.weight q5_K blk.32.ffn_gate.weight iq3_xxs blk.32.ffn_up.weight iq3_xxs -blk.33.ffn_gate.weight iq3_xxs -blk.33.ffn_up.weight iq3_xxs blk.33.attn_k.weight iq3_xxs blk.33.attn_q.weight iq3_xxs blk.33.attn_v.weight q5_K -blk.34.ffn_gate.weight iq3_xxs -blk.34.ffn_up.weight iq3_xxs +blk.33.ffn_gate.weight iq3_xxs +blk.33.ffn_up.weight iq3_xxs blk.34.attn_k.weight iq3_xxs blk.34.attn_q.weight iq3_xxs blk.34.attn_v.weight q5_K +blk.34.ffn_gate.weight iq3_xxs +blk.34.ffn_up.weight iq3_xxs blk.35.attn_k.weight iq3_xxs blk.35.attn_q.weight iq3_xxs blk.35.attn_v.weight q5_K blk.35.ffn_gate.weight iq3_xxs blk.35.ffn_up.weight iq3_xxs -blk.36.ffn_gate.weight iq3_xxs -blk.36.ffn_up.weight iq3_xxs blk.36.attn_k.weight iq3_xxs blk.36.attn_q.weight iq3_xxs blk.36.attn_v.weight q5_K -blk.37.ffn_gate.weight iq3_xxs -blk.37.ffn_up.weight iq3_xxs +blk.36.ffn_gate.weight iq3_xxs +blk.36.ffn_up.weight iq3_xxs blk.37.attn_k.weight iq3_xxs blk.37.attn_q.weight iq3_xxs blk.37.attn_v.weight q5_K -blk.38.ffn_gate.weight iq3_xxs -blk.38.ffn_up.weight iq3_xxs +blk.37.ffn_gate.weight iq3_xxs +blk.37.ffn_up.weight iq3_xxs blk.38.attn_k.weight iq3_xxs blk.38.attn_q.weight iq3_xxs blk.38.attn_v.weight q5_K -blk.39.ffn_gate.weight iq3_xxs -blk.39.ffn_up.weight iq3_xxs +blk.38.ffn_gate.weight iq3_xxs +blk.38.ffn_up.weight iq3_xxs blk.39.attn_k.weight iq3_xxs blk.39.attn_q.weight iq3_xxs blk.39.attn_v.weight q5_K -blk.40.ffn_gate.weight iq3_xxs -blk.40.ffn_up.weight iq3_xxs +blk.39.ffn_gate.weight iq3_xxs +blk.39.ffn_up.weight iq3_xxs blk.40.attn_k.weight iq3_xxs blk.40.attn_q.weight iq3_xxs blk.40.attn_v.weight q5_K -blk.41.ffn_gate.weight iq3_xxs -blk.41.ffn_up.weight iq3_xxs +blk.40.ffn_gate.weight iq3_xxs +blk.40.ffn_up.weight iq3_xxs blk.41.attn_k.weight iq3_xxs blk.41.attn_q.weight iq3_xxs blk.41.attn_v.weight q5_K -blk.42.ffn_gate.weight iq3_xxs -blk.42.ffn_up.weight iq3_xxs +blk.41.ffn_gate.weight iq3_xxs +blk.41.ffn_up.weight iq3_xxs blk.42.attn_k.weight iq3_xxs blk.42.attn_q.weight iq3_xxs blk.42.attn_v.weight q5_K -blk.43.ffn_gate.weight iq3_xxs +blk.42.ffn_gate.weight iq3_xxs +blk.42.ffn_up.weight iq3_xxs blk.43.attn_k.weight iq3_xxs blk.43.attn_q.weight iq3_xxs blk.43.attn_v.weight q5_K +blk.43.ffn_gate.weight iq3_xxs blk.43.ffn_up.weight iq3_xxs -blk.44.ffn_gate.weight iq3_xxs -blk.44.ffn_up.weight iq3_xxs blk.44.attn_k.weight iq3_xxs blk.44.attn_q.weight iq3_xxs blk.44.attn_v.weight q5_K -blk.45.ffn_gate.weight iq3_xxs -blk.45.ffn_up.weight iq3_xxs +blk.44.ffn_gate.weight iq3_xxs +blk.44.ffn_up.weight iq3_xxs blk.45.attn_k.weight iq3_xxs blk.45.attn_q.weight iq3_xxs blk.45.attn_v.weight q5_K +blk.45.ffn_gate.weight iq3_xxs +blk.45.ffn_up.weight iq3_xxs blk.46.attn_k.weight iq3_xxs blk.46.attn_q.weight iq3_xxs blk.46.attn_v.weight q5_K blk.46.ffn_gate.weight iq3_xxs blk.46.ffn_up.weight iq3_xxs -blk.47.ffn_gate.weight iq3_xxs -blk.47.ffn_up.weight iq3_xxs blk.47.attn_k.weight iq3_xxs blk.47.attn_q.weight iq3_xxs blk.47.attn_v.weight q5_K -blk.48.ffn_gate.weight iq3_xxs -blk.48.ffn_up.weight iq3_xxs +blk.47.ffn_gate.weight iq3_xxs +blk.47.ffn_up.weight iq3_xxs blk.48.attn_k.weight iq3_xxs blk.48.attn_q.weight iq3_xxs blk.48.attn_v.weight q5_K +blk.48.ffn_gate.weight iq3_xxs +blk.48.ffn_up.weight iq3_xxs blk.49.attn_k.weight iq3_xxs blk.49.attn_q.weight iq3_xxs blk.49.attn_v.weight q5_K blk.49.ffn_gate.weight iq3_xxs blk.49.ffn_up.weight iq3_xxs -blk.50.ffn_gate.weight iq3_xxs -blk.50.ffn_up.weight iq3_xxs blk.50.attn_k.weight iq3_xxs blk.50.attn_q.weight iq3_xxs blk.50.attn_v.weight q5_K -blk.51.ffn_gate.weight iq3_xxs -blk.51.ffn_up.weight iq3_xxs +blk.50.ffn_gate.weight iq3_xxs +blk.50.ffn_up.weight iq3_xxs blk.51.attn_k.weight iq3_xxs blk.51.attn_q.weight iq3_xxs blk.51.attn_v.weight q5_K -blk.52.ffn_gate.weight iq3_xxs -blk.52.ffn_up.weight iq3_xxs +blk.51.ffn_gate.weight iq3_xxs +blk.51.ffn_up.weight iq3_xxs blk.52.attn_k.weight iq3_xxs blk.52.attn_q.weight iq3_xxs blk.52.attn_v.weight q5_K -blk.53.ffn_gate.weight iq3_xxs -blk.53.ffn_up.weight iq3_xxs +blk.52.ffn_gate.weight iq3_xxs +blk.52.ffn_up.weight iq3_xxs blk.53.attn_k.weight iq3_xxs blk.53.attn_q.weight iq3_xxs blk.53.attn_v.weight q5_K -blk.54.ffn_gate.weight iq3_xxs -blk.54.ffn_up.weight iq3_xxs +blk.53.ffn_gate.weight iq3_xxs +blk.53.ffn_up.weight iq3_xxs blk.54.attn_k.weight iq3_xxs blk.54.attn_q.weight iq3_xxs blk.54.attn_v.weight q5_K -blk.55.ffn_gate.weight iq3_xxs -blk.55.ffn_up.weight iq3_xxs +blk.54.ffn_gate.weight iq3_xxs +blk.54.ffn_up.weight iq3_xxs blk.55.attn_k.weight iq3_xxs blk.55.attn_q.weight iq3_xxs blk.55.attn_v.weight q5_K -blk.56.ffn_gate.weight iq3_xxs -blk.56.ffn_up.weight iq3_xxs +blk.55.ffn_gate.weight iq3_xxs +blk.55.ffn_up.weight iq3_xxs blk.56.attn_k.weight iq3_xxs blk.56.attn_q.weight iq3_xxs blk.56.attn_v.weight q5_K -blk.57.ffn_gate.weight iq3_xxs +blk.56.ffn_gate.weight iq3_xxs +blk.56.ffn_up.weight iq3_xxs blk.57.attn_k.weight iq3_xxs blk.57.attn_q.weight iq3_xxs blk.57.attn_v.weight q5_K +blk.57.ffn_gate.weight iq3_xxs blk.57.ffn_up.weight iq3_xxs -blk.58.ffn_gate.weight iq3_xxs -blk.58.ffn_up.weight iq3_xxs blk.58.attn_k.weight iq3_xxs blk.58.attn_q.weight iq3_xxs blk.58.attn_v.weight q5_K -blk.59.ffn_gate.weight iq3_xxs -blk.59.ffn_up.weight iq3_xxs +blk.58.ffn_gate.weight iq3_xxs +blk.58.ffn_up.weight iq3_xxs blk.59.attn_k.weight iq3_xxs blk.59.attn_q.weight iq3_xxs blk.59.attn_v.weight q5_K +blk.59.ffn_gate.weight iq3_xxs +blk.59.ffn_up.weight iq3_xxs blk.60.attn_k.weight iq3_xxs blk.60.attn_q.weight iq3_xxs blk.60.attn_v.weight q5_K blk.60.ffn_gate.weight iq3_xxs blk.60.ffn_up.weight iq3_xxs -blk.61.ffn_gate.weight iq3_xxs -blk.61.ffn_up.weight iq3_xxs blk.61.attn_k.weight iq3_xxs blk.61.attn_q.weight iq3_xxs blk.61.attn_v.weight q5_K -blk.62.ffn_gate.weight iq3_xxs -blk.62.ffn_up.weight iq3_xxs +blk.61.ffn_gate.weight iq3_xxs +blk.61.ffn_up.weight iq3_xxs blk.62.attn_k.weight iq3_xxs blk.62.attn_q.weight iq3_xxs blk.62.attn_v.weight q5_K +blk.62.ffn_gate.weight iq3_xxs +blk.62.ffn_up.weight iq3_xxs blk.63.attn_k.weight iq3_xxs blk.63.attn_q.weight iq3_xxs blk.63.attn_v.weight q5_K blk.63.ffn_gate.weight iq3_xxs blk.63.ffn_up.weight iq3_xxs -blk.64.ffn_gate.weight iq3_xxs -blk.64.ffn_up.weight iq3_xxs blk.64.attn_k.weight iq3_xxs blk.64.attn_q.weight iq3_xxs blk.64.attn_v.weight q5_K -blk.65.ffn_gate.weight iq3_xxs -blk.65.ffn_up.weight iq3_xxs +blk.64.ffn_gate.weight iq3_xxs +blk.64.ffn_up.weight iq3_xxs blk.65.attn_k.weight iq3_xxs blk.65.attn_q.weight iq3_xxs blk.65.attn_v.weight q5_K -blk.66.ffn_gate.weight iq3_xxs -blk.66.ffn_up.weight iq3_xxs +blk.65.ffn_gate.weight iq3_xxs +blk.65.ffn_up.weight iq3_xxs blk.66.attn_k.weight iq3_xxs blk.66.attn_q.weight iq3_xxs blk.66.attn_v.weight q5_K -blk.67.ffn_gate.weight iq3_xxs -blk.67.ffn_up.weight iq3_xxs +blk.66.ffn_gate.weight iq3_xxs +blk.66.ffn_up.weight iq3_xxs blk.67.attn_k.weight iq3_xxs blk.67.attn_q.weight iq3_xxs blk.67.attn_v.weight q5_K -blk.68.ffn_gate.weight iq3_xxs -blk.68.ffn_up.weight iq3_xxs +blk.67.ffn_gate.weight iq3_xxs +blk.67.ffn_up.weight iq3_xxs blk.68.attn_k.weight iq3_xxs blk.68.attn_q.weight iq3_xxs blk.68.attn_v.weight q5_K -blk.69.ffn_gate.weight iq3_xxs -blk.69.ffn_up.weight iq3_xxs +blk.68.ffn_gate.weight iq3_xxs +blk.68.ffn_up.weight iq3_xxs blk.69.attn_k.weight iq3_xxs blk.69.attn_q.weight iq3_xxs blk.69.attn_v.weight q5_K +blk.69.ffn_gate.weight iq3_xxs +blk.69.ffn_up.weight iq3_xxs blk.70.attn_k.weight iq3_xxs blk.70.attn_q.weight iq3_xxs blk.70.attn_v.weight q5_K @@ -1777,444 +1778,444 @@ blk.78.attn_v.weight q5_K blk.79.attn_k.weight iq3_xxs blk.79.attn_q.weight iq3_xxs blk.79.attn_v.weight q5_K -output.weight q6_K [IQ3_XXS] iq3_xxs +output.weight q5_K token_embd.weight iq3_s -blk.0.ffn_down.weight q4_K blk.0.attn_k.weight iq2_s blk.0.attn_output.weight iq3_s blk.0.attn_q.weight iq2_s blk.0.attn_v.weight q5_K +blk.0.ffn_down.weight q4_K blk.1.attn_k.weight iq2_s blk.1.attn_output.weight iq3_s blk.1.attn_q.weight iq2_s blk.1.attn_v.weight q5_K blk.1.ffn_down.weight q4_K -blk.2.ffn_down.weight q4_K blk.2.attn_k.weight iq2_s blk.2.attn_output.weight iq3_s blk.2.attn_q.weight iq2_s blk.2.attn_v.weight q5_K -blk.3.ffn_down.weight q4_K +blk.2.ffn_down.weight q4_K blk.3.attn_k.weight iq2_s blk.3.attn_output.weight iq3_s blk.3.attn_q.weight iq2_s blk.3.attn_v.weight q5_K +blk.3.ffn_down.weight q4_K blk.4.attn_k.weight iq2_s blk.4.attn_output.weight iq3_s blk.4.attn_q.weight iq2_s blk.4.attn_v.weight q5_K blk.4.ffn_down.weight q4_K -blk.5.ffn_down.weight q4_K blk.5.attn_k.weight iq2_s blk.5.attn_output.weight iq3_s blk.5.attn_q.weight iq2_s blk.5.attn_v.weight q5_K -blk.6.ffn_down.weight q4_K +blk.5.ffn_down.weight q4_K blk.6.attn_k.weight iq2_s blk.6.attn_output.weight iq3_s blk.6.attn_q.weight iq2_s blk.6.attn_v.weight q5_K +blk.6.ffn_down.weight q4_K blk.7.attn_k.weight iq2_s +blk.7.attn_output.weight iq3_s blk.7.attn_q.weight iq2_s blk.7.attn_v.weight q5_K blk.7.ffn_down.weight q4_K -blk.7.attn_output.weight iq3_s -blk.8.ffn_down.weight q4_K blk.8.attn_k.weight iq2_s blk.8.attn_output.weight iq3_s blk.8.attn_q.weight iq2_s blk.8.attn_v.weight q5_K -blk.9.ffn_down.weight q4_K +blk.8.ffn_down.weight q4_K blk.9.attn_k.weight iq2_s blk.9.attn_output.weight iq3_s blk.9.attn_q.weight iq2_s blk.9.attn_v.weight q5_K -blk.10.ffn_down.weight q3_K +blk.9.ffn_down.weight q4_K blk.10.attn_k.weight iq2_s blk.10.attn_output.weight iq3_s blk.10.attn_q.weight iq2_s blk.10.attn_v.weight q5_K -blk.11.ffn_down.weight q3_K +blk.10.ffn_down.weight q3_K blk.11.attn_k.weight iq2_s blk.11.attn_output.weight iq3_s blk.11.attn_q.weight iq2_s blk.11.attn_v.weight q5_K +blk.11.ffn_down.weight q3_K blk.12.attn_k.weight iq2_s blk.12.attn_output.weight iq3_s blk.12.attn_q.weight iq2_s blk.12.attn_v.weight q5_K blk.12.ffn_down.weight q3_K -blk.13.ffn_down.weight q3_K blk.13.attn_k.weight iq2_s blk.13.attn_output.weight iq3_s blk.13.attn_q.weight iq2_s blk.13.attn_v.weight q5_K -blk.14.ffn_down.weight q3_K +blk.13.ffn_down.weight q3_K blk.14.attn_k.weight iq2_s blk.14.attn_output.weight iq3_s blk.14.attn_q.weight iq2_s blk.14.attn_v.weight q5_K +blk.14.ffn_down.weight q3_K blk.15.attn_k.weight iq2_s blk.15.attn_output.weight iq3_s blk.15.attn_q.weight iq2_s blk.15.attn_v.weight q5_K blk.15.ffn_down.weight q3_K -blk.16.ffn_down.weight q3_K blk.16.attn_k.weight iq2_s blk.16.attn_output.weight iq3_s blk.16.attn_q.weight iq2_s blk.16.attn_v.weight q5_K -blk.17.ffn_down.weight q3_K +blk.16.ffn_down.weight q3_K blk.17.attn_k.weight iq2_s blk.17.attn_output.weight iq3_s blk.17.attn_q.weight iq2_s blk.17.attn_v.weight q5_K +blk.17.ffn_down.weight q3_K blk.18.attn_k.weight iq2_s blk.18.attn_output.weight iq3_s blk.18.attn_q.weight iq2_s blk.18.attn_v.weight q5_K blk.18.ffn_down.weight q3_K -blk.19.ffn_down.weight q3_K blk.19.attn_k.weight iq2_s blk.19.attn_output.weight iq3_s blk.19.attn_q.weight iq2_s blk.19.attn_v.weight q5_K -blk.20.ffn_down.weight q3_K +blk.19.ffn_down.weight q3_K blk.20.attn_k.weight iq2_s blk.20.attn_output.weight iq3_s blk.20.attn_q.weight iq2_s blk.20.attn_v.weight q5_K +blk.20.ffn_down.weight q3_K blk.21.attn_k.weight iq2_s +blk.21.attn_output.weight iq3_s blk.21.attn_q.weight iq2_s blk.21.attn_v.weight q5_K blk.21.ffn_down.weight q3_K -blk.21.attn_output.weight iq3_s -blk.22.ffn_down.weight q3_K blk.22.attn_k.weight iq2_s blk.22.attn_output.weight iq3_s blk.22.attn_q.weight iq2_s blk.22.attn_v.weight q5_K -blk.23.ffn_down.weight q3_K +blk.22.ffn_down.weight q3_K blk.23.attn_k.weight iq2_s blk.23.attn_output.weight iq3_s blk.23.attn_q.weight iq2_s blk.23.attn_v.weight q5_K -blk.24.ffn_down.weight q3_K +blk.23.ffn_down.weight q3_K blk.24.attn_k.weight iq2_s blk.24.attn_output.weight iq3_s blk.24.attn_q.weight iq2_s blk.24.attn_v.weight q5_K -blk.25.ffn_down.weight q3_K +blk.24.ffn_down.weight q3_K blk.25.attn_k.weight iq2_s blk.25.attn_output.weight iq3_s blk.25.attn_q.weight iq2_s blk.25.attn_v.weight q5_K +blk.25.ffn_down.weight q3_K blk.26.attn_k.weight iq2_s blk.26.attn_output.weight iq3_s blk.26.attn_q.weight iq2_s blk.26.attn_v.weight q5_K blk.26.ffn_down.weight q3_K -blk.27.ffn_down.weight q3_K blk.27.attn_k.weight iq2_s blk.27.attn_output.weight iq3_s blk.27.attn_q.weight iq2_s blk.27.attn_v.weight q5_K -blk.28.ffn_down.weight q3_K +blk.27.ffn_down.weight q3_K blk.28.attn_k.weight iq2_s blk.28.attn_output.weight iq3_s blk.28.attn_q.weight iq2_s blk.28.attn_v.weight q5_K +blk.28.ffn_down.weight q3_K blk.29.attn_k.weight iq2_s blk.29.attn_output.weight iq3_s blk.29.attn_q.weight iq2_s blk.29.attn_v.weight q5_K blk.29.ffn_down.weight q3_K -blk.30.ffn_down.weight q3_K blk.30.attn_k.weight iq2_s blk.30.attn_output.weight iq3_s blk.30.attn_q.weight iq2_s blk.30.attn_v.weight q5_K -blk.31.ffn_down.weight q3_K +blk.30.ffn_down.weight q3_K blk.31.attn_k.weight iq2_s blk.31.attn_output.weight iq3_s blk.31.attn_q.weight iq2_s blk.31.attn_v.weight q5_K +blk.31.ffn_down.weight q3_K blk.32.attn_k.weight iq2_s blk.32.attn_output.weight iq3_s blk.32.attn_q.weight iq2_s blk.32.attn_v.weight q5_K blk.32.ffn_down.weight q3_K -blk.33.ffn_down.weight q3_K blk.33.attn_k.weight iq2_s blk.33.attn_output.weight iq3_s blk.33.attn_q.weight iq2_s blk.33.attn_v.weight q5_K -blk.34.ffn_down.weight q3_K +blk.33.ffn_down.weight q3_K blk.34.attn_k.weight iq2_s blk.34.attn_output.weight iq3_s blk.34.attn_q.weight iq2_s blk.34.attn_v.weight q5_K +blk.34.ffn_down.weight q3_K blk.35.attn_k.weight iq2_s +blk.35.attn_output.weight iq3_s blk.35.attn_q.weight iq2_s blk.35.attn_v.weight q5_K blk.35.ffn_down.weight q3_K -blk.35.attn_output.weight iq3_s -blk.36.ffn_down.weight q3_K blk.36.attn_k.weight iq2_s blk.36.attn_output.weight iq3_s blk.36.attn_q.weight iq2_s blk.36.attn_v.weight q5_K -blk.37.ffn_down.weight q3_K +blk.36.ffn_down.weight q3_K blk.37.attn_k.weight iq2_s blk.37.attn_output.weight iq3_s blk.37.attn_q.weight iq2_s blk.37.attn_v.weight q5_K -blk.38.ffn_down.weight q3_K +blk.37.ffn_down.weight q3_K blk.38.attn_k.weight iq2_s blk.38.attn_output.weight iq3_s blk.38.attn_q.weight iq2_s blk.38.attn_v.weight q5_K -blk.39.ffn_down.weight q3_K +blk.38.ffn_down.weight q3_K blk.39.attn_k.weight iq2_s blk.39.attn_output.weight iq3_s blk.39.attn_q.weight iq2_s blk.39.attn_v.weight q5_K +blk.39.ffn_down.weight q3_K blk.40.attn_k.weight iq2_s blk.40.attn_output.weight iq3_s blk.40.attn_q.weight iq2_s blk.40.attn_v.weight q5_K blk.40.ffn_down.weight q3_K -blk.41.ffn_down.weight q3_K blk.41.attn_k.weight iq2_s blk.41.attn_output.weight iq3_s blk.41.attn_q.weight iq2_s blk.41.attn_v.weight q5_K -blk.42.ffn_down.weight q3_K +blk.41.ffn_down.weight q3_K blk.42.attn_k.weight iq2_s blk.42.attn_output.weight iq3_s blk.42.attn_q.weight iq2_s blk.42.attn_v.weight q5_K +blk.42.ffn_down.weight q3_K blk.43.attn_k.weight iq2_s blk.43.attn_output.weight iq3_s blk.43.attn_q.weight iq2_s blk.43.attn_v.weight q5_K blk.43.ffn_down.weight q3_K -blk.44.ffn_down.weight q3_K blk.44.attn_k.weight iq2_s blk.44.attn_output.weight iq3_s blk.44.attn_q.weight iq2_s blk.44.attn_v.weight q5_K -blk.45.ffn_down.weight q3_K +blk.44.ffn_down.weight q3_K blk.45.attn_k.weight iq2_s blk.45.attn_output.weight iq3_s blk.45.attn_q.weight iq2_s blk.45.attn_v.weight q5_K +blk.45.ffn_down.weight q3_K blk.46.attn_k.weight iq2_s blk.46.attn_output.weight iq3_s blk.46.attn_q.weight iq2_s blk.46.attn_v.weight q5_K blk.46.ffn_down.weight q3_K -blk.47.ffn_down.weight q3_K blk.47.attn_k.weight iq2_s blk.47.attn_output.weight iq3_s blk.47.attn_q.weight iq2_s blk.47.attn_v.weight q5_K -blk.48.ffn_down.weight q3_K +blk.47.ffn_down.weight q3_K blk.48.attn_k.weight iq2_s blk.48.attn_output.weight iq3_s blk.48.attn_q.weight iq2_s blk.48.attn_v.weight q5_K +blk.48.ffn_down.weight q3_K blk.49.attn_k.weight iq2_s +blk.49.attn_output.weight iq3_s blk.49.attn_q.weight iq2_s blk.49.attn_v.weight q5_K blk.49.ffn_down.weight q3_K -blk.49.attn_output.weight iq3_s -blk.50.ffn_down.weight q3_K blk.50.attn_k.weight iq2_s blk.50.attn_output.weight iq3_s blk.50.attn_q.weight iq2_s blk.50.attn_v.weight q5_K -blk.51.ffn_down.weight q3_K +blk.50.ffn_down.weight q3_K blk.51.attn_k.weight iq2_s blk.51.attn_output.weight iq3_s blk.51.attn_q.weight iq2_s blk.51.attn_v.weight q5_K -blk.52.ffn_down.weight q3_K +blk.51.ffn_down.weight q3_K blk.52.attn_k.weight iq2_s blk.52.attn_output.weight iq3_s blk.52.attn_q.weight iq2_s blk.52.attn_v.weight q5_K -blk.53.ffn_down.weight q3_K +blk.52.ffn_down.weight q3_K blk.53.attn_k.weight iq2_s blk.53.attn_output.weight iq3_s blk.53.attn_q.weight iq2_s blk.53.attn_v.weight q5_K +blk.53.ffn_down.weight q3_K blk.54.attn_k.weight iq2_s blk.54.attn_output.weight iq3_s blk.54.attn_q.weight iq2_s blk.54.attn_v.weight q5_K blk.54.ffn_down.weight q3_K -blk.55.ffn_down.weight q3_K blk.55.attn_k.weight iq2_s blk.55.attn_output.weight iq3_s blk.55.attn_q.weight iq2_s blk.55.attn_v.weight q5_K -blk.56.ffn_down.weight q3_K +blk.55.ffn_down.weight q3_K blk.56.attn_k.weight iq2_s blk.56.attn_output.weight iq3_s blk.56.attn_q.weight iq2_s blk.56.attn_v.weight q5_K +blk.56.ffn_down.weight q3_K blk.57.attn_k.weight iq2_s blk.57.attn_output.weight iq3_s blk.57.attn_q.weight iq2_s blk.57.attn_v.weight q5_K blk.57.ffn_down.weight q3_K -blk.58.ffn_down.weight q3_K blk.58.attn_k.weight iq2_s blk.58.attn_output.weight iq3_s blk.58.attn_q.weight iq2_s blk.58.attn_v.weight q5_K -blk.59.ffn_down.weight q3_K +blk.58.ffn_down.weight q3_K blk.59.attn_k.weight iq2_s blk.59.attn_output.weight iq3_s blk.59.attn_q.weight iq2_s blk.59.attn_v.weight q5_K +blk.59.ffn_down.weight q3_K blk.60.attn_k.weight iq2_s blk.60.attn_output.weight iq3_s blk.60.attn_q.weight iq2_s blk.60.attn_v.weight q5_K blk.60.ffn_down.weight q3_K -blk.61.ffn_down.weight q3_K blk.61.attn_k.weight iq2_s blk.61.attn_output.weight iq3_s blk.61.attn_q.weight iq2_s blk.61.attn_v.weight q5_K -blk.62.ffn_down.weight q3_K +blk.61.ffn_down.weight q3_K blk.62.attn_k.weight iq2_s blk.62.attn_output.weight iq3_s blk.62.attn_q.weight iq2_s blk.62.attn_v.weight q5_K +blk.62.ffn_down.weight q3_K blk.63.attn_k.weight iq2_s +blk.63.attn_output.weight iq3_s blk.63.attn_q.weight iq2_s blk.63.attn_v.weight q5_K blk.63.ffn_down.weight q3_K -blk.63.attn_output.weight iq3_s -blk.64.ffn_down.weight q3_K blk.64.attn_k.weight iq2_s blk.64.attn_output.weight iq3_s blk.64.attn_q.weight iq2_s blk.64.attn_v.weight q5_K -blk.65.ffn_down.weight q3_K +blk.64.ffn_down.weight q3_K blk.65.attn_k.weight iq2_s blk.65.attn_output.weight iq3_s blk.65.attn_q.weight iq2_s blk.65.attn_v.weight q5_K -blk.66.ffn_down.weight q3_K +blk.65.ffn_down.weight q3_K blk.66.attn_k.weight iq2_s blk.66.attn_output.weight iq3_s blk.66.attn_q.weight iq2_s blk.66.attn_v.weight q5_K -blk.67.ffn_down.weight q3_K +blk.66.ffn_down.weight q3_K blk.67.attn_k.weight iq2_s blk.67.attn_output.weight iq3_s blk.67.attn_q.weight iq2_s blk.67.attn_v.weight q5_K +blk.67.ffn_down.weight q3_K blk.68.attn_k.weight iq2_s blk.68.attn_output.weight iq3_s blk.68.attn_q.weight iq2_s blk.68.attn_v.weight q5_K blk.68.ffn_down.weight q3_K -blk.69.ffn_down.weight q3_K blk.69.attn_k.weight iq2_s blk.69.attn_output.weight iq3_s blk.69.attn_q.weight iq2_s blk.69.attn_v.weight q5_K -blk.70.ffn_down.weight q3_K +blk.69.ffn_down.weight q3_K blk.70.attn_k.weight iq2_s blk.70.attn_output.weight iq3_s blk.70.attn_q.weight iq2_s blk.70.attn_v.weight q5_K +blk.70.ffn_down.weight q3_K blk.71.attn_k.weight iq2_s blk.71.attn_output.weight iq3_s blk.71.attn_q.weight iq2_s blk.71.attn_v.weight q5_K blk.71.ffn_down.weight q3_K -blk.72.ffn_down.weight q3_K blk.72.attn_k.weight iq2_s blk.72.attn_output.weight iq3_s blk.72.attn_q.weight iq2_s blk.72.attn_v.weight q5_K -blk.73.ffn_down.weight q3_K +blk.72.ffn_down.weight q3_K blk.73.attn_k.weight iq2_s blk.73.attn_output.weight iq3_s blk.73.attn_q.weight iq2_s blk.73.attn_v.weight q5_K +blk.73.ffn_down.weight q3_K blk.74.attn_k.weight iq2_s blk.74.attn_output.weight iq3_s blk.74.attn_q.weight iq2_s blk.74.attn_v.weight q5_K blk.74.ffn_down.weight q3_K -blk.75.ffn_down.weight q3_K blk.75.attn_k.weight iq2_s blk.75.attn_output.weight iq3_s blk.75.attn_q.weight iq2_s blk.75.attn_v.weight q5_K -blk.76.ffn_down.weight q3_K +blk.75.ffn_down.weight q3_K blk.76.attn_k.weight iq2_s blk.76.attn_output.weight iq3_s blk.76.attn_q.weight iq2_s blk.76.attn_v.weight q5_K +blk.76.ffn_down.weight q3_K blk.77.attn_k.weight iq2_s +blk.77.attn_output.weight iq3_s blk.77.attn_q.weight iq2_s blk.77.attn_v.weight q5_K blk.77.ffn_down.weight q3_K -blk.77.attn_output.weight iq3_s -blk.78.ffn_down.weight q3_K blk.78.attn_k.weight iq2_s blk.78.attn_output.weight iq3_s blk.78.attn_q.weight iq2_s blk.78.attn_v.weight q5_K -blk.79.ffn_down.weight q3_K +blk.78.ffn_down.weight q3_K blk.79.attn_k.weight iq2_s blk.79.attn_output.weight iq3_s blk.79.attn_q.weight iq2_s blk.79.attn_v.weight q5_K -output.weight q5_K +blk.79.ffn_down.weight q3_K [IQ1_S] iq1_s +output.weight q5_K token_embd.weight q2_K -blk.0.ffn_down.weight q2_K blk.0.attn_output.weight iq2_xxs blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q2_K blk.1.attn_output.weight iq2_xxs blk.1.attn_v.weight q4_K blk.1.ffn_down.weight q2_K -blk.2.ffn_down.weight q2_K blk.2.attn_output.weight iq2_xxs blk.2.attn_v.weight q4_K -blk.3.ffn_down.weight q2_K +blk.2.ffn_down.weight q2_K blk.3.attn_output.weight iq2_xxs blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight q2_K blk.4.attn_output.weight iq2_xxs blk.4.attn_v.weight q4_K blk.4.ffn_down.weight q2_K -blk.5.ffn_down.weight q2_K blk.5.attn_output.weight iq2_xxs blk.5.attn_v.weight q4_K -blk.6.ffn_down.weight q2_K +blk.5.ffn_down.weight q2_K blk.6.attn_output.weight iq2_xxs blk.6.attn_v.weight q4_K +blk.6.ffn_down.weight q2_K +blk.7.attn_output.weight iq2_xxs blk.7.attn_v.weight q4_K blk.7.ffn_down.weight q2_K -blk.7.attn_output.weight iq2_xxs -blk.8.ffn_down.weight q2_K blk.8.attn_output.weight iq2_xxs blk.8.attn_v.weight q4_K -blk.9.ffn_down.weight q2_K +blk.8.ffn_down.weight q2_K blk.9.attn_output.weight iq2_xxs blk.9.attn_v.weight q4_K +blk.9.ffn_down.weight q2_K blk.10.attn_output.weight iq2_xxs blk.10.attn_v.weight q4_K blk.11.attn_output.weight iq2_xxs @@ -2237,8 +2238,8 @@ blk.19.attn_output.weight iq2_xxs blk.19.attn_v.weight q4_K blk.20.attn_output.weight iq2_xxs blk.20.attn_v.weight q4_K -blk.21.attn_v.weight q4_K blk.21.attn_output.weight iq2_xxs +blk.21.attn_v.weight q4_K blk.22.attn_output.weight iq2_xxs blk.22.attn_v.weight q4_K blk.23.attn_output.weight iq2_xxs @@ -2265,8 +2266,8 @@ blk.33.attn_output.weight iq2_xxs blk.33.attn_v.weight q4_K blk.34.attn_output.weight iq2_xxs blk.34.attn_v.weight q4_K -blk.35.attn_v.weight q4_K blk.35.attn_output.weight iq2_xxs +blk.35.attn_v.weight q4_K blk.36.attn_output.weight iq2_xxs blk.36.attn_v.weight q4_K blk.37.attn_output.weight iq2_xxs @@ -2293,8 +2294,8 @@ blk.47.attn_output.weight iq2_xxs blk.47.attn_v.weight q4_K blk.48.attn_output.weight iq2_xxs blk.48.attn_v.weight q4_K -blk.49.attn_v.weight q4_K blk.49.attn_output.weight iq2_xxs +blk.49.attn_v.weight q4_K blk.50.attn_output.weight iq2_xxs blk.50.attn_v.weight q4_K blk.51.attn_output.weight iq2_xxs @@ -2321,8 +2322,8 @@ blk.61.attn_output.weight iq2_xxs blk.61.attn_v.weight q4_K blk.62.attn_output.weight iq2_xxs blk.62.attn_v.weight q4_K -blk.63.attn_v.weight q4_K blk.63.attn_output.weight iq2_xxs +blk.63.attn_v.weight q4_K blk.64.attn_output.weight iq2_xxs blk.64.attn_v.weight q4_K blk.65.attn_output.weight iq2_xxs @@ -2349,35 +2350,35 @@ blk.75.attn_output.weight iq2_xxs blk.75.attn_v.weight q4_K blk.76.attn_output.weight iq2_xxs blk.76.attn_v.weight q4_K -blk.77.attn_v.weight q4_K blk.77.attn_output.weight iq2_xxs +blk.77.attn_v.weight q4_K blk.78.attn_output.weight iq2_xxs blk.78.attn_v.weight q4_K blk.79.attn_output.weight iq2_xxs blk.79.attn_v.weight q4_K -output.weight q5_K [IQ4_NL] iq4_nl -blk.0.ffn_down.weight q5_K +output.weight q6_K blk.0.attn_v.weight q5_K +blk.0.ffn_down.weight q5_K blk.1.attn_v.weight q5_K blk.1.ffn_down.weight q5_K -blk.2.ffn_down.weight q5_K blk.2.attn_v.weight q5_K -blk.3.ffn_down.weight q5_K +blk.2.ffn_down.weight q5_K blk.3.attn_v.weight q5_K +blk.3.ffn_down.weight q5_K blk.4.attn_v.weight q5_K blk.4.ffn_down.weight q5_K -blk.5.ffn_down.weight q5_K blk.5.attn_v.weight q5_K -blk.6.ffn_down.weight q5_K +blk.5.ffn_down.weight q5_K blk.6.attn_v.weight q5_K +blk.6.ffn_down.weight q5_K blk.7.attn_v.weight q5_K blk.7.ffn_down.weight q5_K -blk.8.ffn_down.weight q5_K blk.8.attn_v.weight q5_K -blk.9.ffn_down.weight q5_K +blk.8.ffn_down.weight q5_K blk.9.attn_v.weight q5_K +blk.9.ffn_down.weight q5_K blk.10.attn_v.weight q5_K blk.11.attn_v.weight q5_K blk.12.attn_v.weight q5_K @@ -2448,9 +2449,9 @@ blk.76.attn_v.weight q5_K blk.77.attn_v.weight q5_K blk.78.attn_v.weight q5_K blk.79.attn_v.weight q5_K -output.weight q6_K [IQ3_S] iq3_s +output.weight q6_K blk.0.attn_v.weight q5_K blk.1.attn_v.weight q5_K blk.2.attn_v.weight q5_K @@ -2531,39 +2532,39 @@ blk.76.attn_v.weight q5_K blk.77.attn_v.weight q5_K blk.78.attn_v.weight q5_K blk.79.attn_v.weight q5_K -output.weight q6_K [IQ3_M] iq3_s -blk.0.ffn_down.weight q4_K +output.weight q6_K blk.0.attn_output.weight q4_K blk.0.attn_v.weight q5_K +blk.0.ffn_down.weight q4_K blk.1.attn_output.weight q4_K blk.1.attn_v.weight q5_K blk.1.ffn_down.weight q4_K -blk.2.ffn_down.weight q4_K blk.2.attn_output.weight q4_K blk.2.attn_v.weight q5_K -blk.3.ffn_down.weight q4_K +blk.2.ffn_down.weight q4_K blk.3.attn_output.weight q4_K blk.3.attn_v.weight q5_K +blk.3.ffn_down.weight q4_K blk.4.attn_output.weight q4_K blk.4.attn_v.weight q5_K blk.4.ffn_down.weight q4_K -blk.5.ffn_down.weight q4_K blk.5.attn_output.weight q4_K blk.5.attn_v.weight q5_K -blk.6.ffn_down.weight q4_K +blk.5.ffn_down.weight q4_K blk.6.attn_output.weight q4_K blk.6.attn_v.weight q5_K +blk.6.ffn_down.weight q4_K +blk.7.attn_output.weight q4_K blk.7.attn_v.weight q5_K blk.7.ffn_down.weight q4_K -blk.7.attn_output.weight q4_K -blk.8.ffn_down.weight q4_K blk.8.attn_output.weight q4_K blk.8.attn_v.weight q5_K -blk.9.ffn_down.weight q4_K +blk.8.ffn_down.weight q4_K blk.9.attn_output.weight q4_K blk.9.attn_v.weight q5_K +blk.9.ffn_down.weight q4_K blk.10.attn_output.weight q4_K blk.10.attn_v.weight q5_K blk.11.attn_output.weight q4_K @@ -2586,8 +2587,8 @@ blk.19.attn_output.weight q4_K blk.19.attn_v.weight q5_K blk.20.attn_output.weight q4_K blk.20.attn_v.weight q5_K -blk.21.attn_v.weight q5_K blk.21.attn_output.weight q4_K +blk.21.attn_v.weight q5_K blk.22.attn_output.weight q4_K blk.22.attn_v.weight q5_K blk.23.attn_output.weight q4_K @@ -2614,8 +2615,8 @@ blk.33.attn_output.weight q4_K blk.33.attn_v.weight q5_K blk.34.attn_output.weight q4_K blk.34.attn_v.weight q5_K -blk.35.attn_v.weight q5_K blk.35.attn_output.weight q4_K +blk.35.attn_v.weight q5_K blk.36.attn_output.weight q4_K blk.36.attn_v.weight q5_K blk.37.attn_output.weight q4_K @@ -2642,8 +2643,8 @@ blk.47.attn_output.weight q4_K blk.47.attn_v.weight q5_K blk.48.attn_output.weight q4_K blk.48.attn_v.weight q5_K -blk.49.attn_v.weight q5_K blk.49.attn_output.weight q4_K +blk.49.attn_v.weight q5_K blk.50.attn_output.weight q4_K blk.50.attn_v.weight q5_K blk.51.attn_output.weight q4_K @@ -2670,8 +2671,8 @@ blk.61.attn_output.weight q4_K blk.61.attn_v.weight q5_K blk.62.attn_output.weight q4_K blk.62.attn_v.weight q5_K -blk.63.attn_v.weight q5_K blk.63.attn_output.weight q4_K +blk.63.attn_v.weight q5_K blk.64.attn_output.weight q4_K blk.64.attn_v.weight q5_K blk.65.attn_output.weight q4_K @@ -2698,46 +2699,46 @@ blk.75.attn_output.weight q4_K blk.75.attn_v.weight q5_K blk.76.attn_output.weight q4_K blk.76.attn_v.weight q5_K -blk.77.attn_v.weight q5_K blk.77.attn_output.weight q4_K +blk.77.attn_v.weight q5_K blk.78.attn_output.weight q4_K blk.78.attn_v.weight q5_K blk.79.attn_output.weight q4_K blk.79.attn_v.weight q5_K -output.weight q6_K [IQ2_S] iq2_xs +output.weight q5_K token_embd.weight iq3_s -blk.0.ffn_down.weight iq3_s blk.0.attn_output.weight iq3_s blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight iq3_s blk.1.attn_output.weight iq3_s blk.1.attn_v.weight q4_K blk.1.ffn_down.weight iq3_s -blk.2.ffn_down.weight iq3_s blk.2.attn_output.weight iq3_s blk.2.attn_v.weight q4_K -blk.3.ffn_down.weight iq3_s +blk.2.ffn_down.weight iq3_s blk.3.attn_output.weight iq3_s blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight iq3_s blk.4.attn_output.weight iq3_s blk.4.attn_v.weight q4_K blk.4.ffn_down.weight iq3_s -blk.5.ffn_down.weight iq3_s blk.5.attn_output.weight iq3_s blk.5.attn_v.weight q4_K -blk.6.ffn_down.weight iq3_s +blk.5.ffn_down.weight iq3_s blk.6.attn_output.weight iq3_s blk.6.attn_v.weight q4_K +blk.6.ffn_down.weight iq3_s +blk.7.attn_output.weight iq3_s blk.7.attn_v.weight q4_K blk.7.ffn_down.weight iq3_s -blk.7.attn_output.weight iq3_s -blk.8.ffn_down.weight iq3_s blk.8.attn_output.weight iq3_s blk.8.attn_v.weight q4_K -blk.9.ffn_down.weight iq3_s +blk.8.ffn_down.weight iq3_s blk.9.attn_output.weight iq3_s blk.9.attn_v.weight q4_K +blk.9.ffn_down.weight iq3_s blk.10.attn_output.weight iq3_s blk.10.attn_v.weight q4_K blk.11.attn_output.weight iq3_s @@ -2760,8 +2761,8 @@ blk.19.attn_output.weight iq3_s blk.19.attn_v.weight q4_K blk.20.attn_output.weight iq3_s blk.20.attn_v.weight q4_K -blk.21.attn_v.weight q4_K blk.21.attn_output.weight iq3_s +blk.21.attn_v.weight q4_K blk.22.attn_output.weight iq3_s blk.22.attn_v.weight q4_K blk.23.attn_output.weight iq3_s @@ -2788,8 +2789,8 @@ blk.33.attn_output.weight iq3_s blk.33.attn_v.weight q4_K blk.34.attn_output.weight iq3_s blk.34.attn_v.weight q4_K -blk.35.attn_v.weight q4_K blk.35.attn_output.weight iq3_s +blk.35.attn_v.weight q4_K blk.36.attn_output.weight iq3_s blk.36.attn_v.weight q4_K blk.37.attn_output.weight iq3_s @@ -2816,8 +2817,8 @@ blk.47.attn_output.weight iq3_s blk.47.attn_v.weight q4_K blk.48.attn_output.weight iq3_s blk.48.attn_v.weight q4_K -blk.49.attn_v.weight q4_K blk.49.attn_output.weight iq3_s +blk.49.attn_v.weight q4_K blk.50.attn_output.weight iq3_s blk.50.attn_v.weight q4_K blk.51.attn_output.weight iq3_s @@ -2844,8 +2845,8 @@ blk.61.attn_output.weight iq3_s blk.61.attn_v.weight q4_K blk.62.attn_output.weight iq3_s blk.62.attn_v.weight q4_K -blk.63.attn_v.weight q4_K blk.63.attn_output.weight iq3_s +blk.63.attn_v.weight q4_K blk.64.attn_output.weight iq3_s blk.64.attn_v.weight q4_K blk.65.attn_output.weight iq3_s @@ -2872,46 +2873,46 @@ blk.75.attn_output.weight iq3_s blk.75.attn_v.weight q4_K blk.76.attn_output.weight iq3_s blk.76.attn_v.weight q4_K -blk.77.attn_v.weight q4_K blk.77.attn_output.weight iq3_s +blk.77.attn_v.weight q4_K blk.78.attn_output.weight iq3_s blk.78.attn_v.weight q4_K blk.79.attn_output.weight iq3_s blk.79.attn_v.weight q4_K -output.weight q5_K [IQ2_M] iq2_s +output.weight q5_K token_embd.weight iq3_s -blk.0.ffn_down.weight iq3_s blk.0.attn_output.weight iq3_s blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight iq3_s blk.1.attn_output.weight iq3_s blk.1.attn_v.weight q4_K blk.1.ffn_down.weight iq3_s -blk.2.ffn_down.weight iq3_s blk.2.attn_output.weight iq3_s blk.2.attn_v.weight q4_K -blk.3.ffn_down.weight iq3_s +blk.2.ffn_down.weight iq3_s blk.3.attn_output.weight iq3_s blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight iq3_s blk.4.attn_output.weight iq3_s blk.4.attn_v.weight q4_K blk.4.ffn_down.weight iq3_s -blk.5.ffn_down.weight iq3_s blk.5.attn_output.weight iq3_s blk.5.attn_v.weight q4_K -blk.6.ffn_down.weight iq3_s +blk.5.ffn_down.weight iq3_s blk.6.attn_output.weight iq3_s blk.6.attn_v.weight q4_K +blk.6.ffn_down.weight iq3_s +blk.7.attn_output.weight iq3_s blk.7.attn_v.weight q4_K blk.7.ffn_down.weight iq3_s -blk.7.attn_output.weight iq3_s -blk.8.ffn_down.weight iq3_s blk.8.attn_output.weight iq3_s blk.8.attn_v.weight q4_K -blk.9.ffn_down.weight iq3_s +blk.8.ffn_down.weight iq3_s blk.9.attn_output.weight iq3_s blk.9.attn_v.weight q4_K +blk.9.ffn_down.weight iq3_s blk.10.attn_output.weight iq3_s blk.10.attn_v.weight q4_K blk.11.attn_output.weight iq3_s @@ -2934,8 +2935,8 @@ blk.19.attn_output.weight iq3_s blk.19.attn_v.weight q4_K blk.20.attn_output.weight iq3_s blk.20.attn_v.weight q4_K -blk.21.attn_v.weight q4_K blk.21.attn_output.weight iq3_s +blk.21.attn_v.weight q4_K blk.22.attn_output.weight iq3_s blk.22.attn_v.weight q4_K blk.23.attn_output.weight iq3_s @@ -2962,8 +2963,8 @@ blk.33.attn_output.weight iq3_s blk.33.attn_v.weight q4_K blk.34.attn_output.weight iq3_s blk.34.attn_v.weight q4_K -blk.35.attn_v.weight q4_K blk.35.attn_output.weight iq3_s +blk.35.attn_v.weight q4_K blk.36.attn_output.weight iq3_s blk.36.attn_v.weight q4_K blk.37.attn_output.weight iq3_s @@ -2990,8 +2991,8 @@ blk.47.attn_output.weight iq3_s blk.47.attn_v.weight q4_K blk.48.attn_output.weight iq3_s blk.48.attn_v.weight q4_K -blk.49.attn_v.weight q4_K blk.49.attn_output.weight iq3_s +blk.49.attn_v.weight q4_K blk.50.attn_output.weight iq3_s blk.50.attn_v.weight q4_K blk.51.attn_output.weight iq3_s @@ -3018,8 +3019,8 @@ blk.61.attn_output.weight iq3_s blk.61.attn_v.weight q4_K blk.62.attn_output.weight iq3_s blk.62.attn_v.weight q4_K -blk.63.attn_v.weight q4_K blk.63.attn_output.weight iq3_s +blk.63.attn_v.weight q4_K blk.64.attn_output.weight iq3_s blk.64.attn_v.weight q4_K blk.65.attn_output.weight iq3_s @@ -3046,35 +3047,35 @@ blk.75.attn_output.weight iq3_s blk.75.attn_v.weight q4_K blk.76.attn_output.weight iq3_s blk.76.attn_v.weight q4_K -blk.77.attn_v.weight q4_K blk.77.attn_output.weight iq3_s +blk.77.attn_v.weight q4_K blk.78.attn_output.weight iq3_s blk.78.attn_v.weight q4_K blk.79.attn_output.weight iq3_s blk.79.attn_v.weight q4_K -output.weight q5_K [IQ4_XS] iq4_xs -blk.0.ffn_down.weight q5_K +output.weight q6_K blk.0.attn_v.weight q5_K +blk.0.ffn_down.weight q5_K blk.1.attn_v.weight q5_K blk.1.ffn_down.weight q5_K -blk.2.ffn_down.weight q5_K blk.2.attn_v.weight q5_K -blk.3.ffn_down.weight q5_K +blk.2.ffn_down.weight q5_K blk.3.attn_v.weight q5_K +blk.3.ffn_down.weight q5_K blk.4.attn_v.weight q5_K blk.4.ffn_down.weight q5_K -blk.5.ffn_down.weight q5_K blk.5.attn_v.weight q5_K -blk.6.ffn_down.weight q5_K +blk.5.ffn_down.weight q5_K blk.6.attn_v.weight q5_K +blk.6.ffn_down.weight q5_K blk.7.attn_v.weight q5_K blk.7.ffn_down.weight q5_K -blk.8.ffn_down.weight q5_K blk.8.attn_v.weight q5_K -blk.9.ffn_down.weight q5_K +blk.8.ffn_down.weight q5_K blk.9.attn_v.weight q5_K +blk.9.ffn_down.weight q5_K blk.10.attn_v.weight q5_K blk.11.attn_v.weight q5_K blk.12.attn_v.weight q5_K @@ -3145,40 +3146,40 @@ blk.76.attn_v.weight q5_K blk.77.attn_v.weight q5_K blk.78.attn_v.weight q5_K blk.79.attn_v.weight q5_K -output.weight q6_K [IQ1_M] iq1_m +output.weight q5_K token_embd.weight q2_K -blk.0.ffn_down.weight q2_K blk.0.attn_output.weight iq2_xxs blk.0.attn_v.weight q4_K +blk.0.ffn_down.weight q2_K blk.1.attn_output.weight iq2_xxs blk.1.attn_v.weight q4_K blk.1.ffn_down.weight q2_K -blk.2.ffn_down.weight q2_K blk.2.attn_output.weight iq2_xxs blk.2.attn_v.weight q4_K -blk.3.ffn_down.weight q2_K +blk.2.ffn_down.weight q2_K blk.3.attn_output.weight iq2_xxs blk.3.attn_v.weight q4_K +blk.3.ffn_down.weight q2_K blk.4.attn_output.weight iq2_xxs blk.4.attn_v.weight q4_K blk.4.ffn_down.weight q2_K -blk.5.ffn_down.weight q2_K blk.5.attn_output.weight iq2_xxs blk.5.attn_v.weight q4_K -blk.6.ffn_down.weight q2_K +blk.5.ffn_down.weight q2_K blk.6.attn_output.weight iq2_xxs blk.6.attn_v.weight q4_K +blk.6.ffn_down.weight q2_K +blk.7.attn_output.weight iq2_xxs blk.7.attn_v.weight q4_K blk.7.ffn_down.weight q2_K -blk.7.attn_output.weight iq2_xxs -blk.8.ffn_down.weight q2_K blk.8.attn_output.weight iq2_xxs blk.8.attn_v.weight q4_K -blk.9.ffn_down.weight q2_K +blk.8.ffn_down.weight q2_K blk.9.attn_output.weight iq2_xxs blk.9.attn_v.weight q4_K +blk.9.ffn_down.weight q2_K blk.10.attn_output.weight iq2_xxs blk.10.attn_v.weight q4_K blk.11.attn_output.weight iq2_xxs @@ -3201,8 +3202,8 @@ blk.19.attn_output.weight iq2_xxs blk.19.attn_v.weight q4_K blk.20.attn_output.weight iq2_xxs blk.20.attn_v.weight q4_K -blk.21.attn_v.weight q4_K blk.21.attn_output.weight iq2_xxs +blk.21.attn_v.weight q4_K blk.22.attn_output.weight iq2_xxs blk.22.attn_v.weight q4_K blk.23.attn_output.weight iq2_xxs @@ -3229,8 +3230,8 @@ blk.33.attn_output.weight iq2_xxs blk.33.attn_v.weight q4_K blk.34.attn_output.weight iq2_xxs blk.34.attn_v.weight q4_K -blk.35.attn_v.weight q4_K blk.35.attn_output.weight iq2_xxs +blk.35.attn_v.weight q4_K blk.36.attn_output.weight iq2_xxs blk.36.attn_v.weight q4_K blk.37.attn_output.weight iq2_xxs @@ -3257,8 +3258,8 @@ blk.47.attn_output.weight iq2_xxs blk.47.attn_v.weight q4_K blk.48.attn_output.weight iq2_xxs blk.48.attn_v.weight q4_K -blk.49.attn_v.weight q4_K blk.49.attn_output.weight iq2_xxs +blk.49.attn_v.weight q4_K blk.50.attn_output.weight iq2_xxs blk.50.attn_v.weight q4_K blk.51.attn_output.weight iq2_xxs @@ -3285,8 +3286,8 @@ blk.61.attn_output.weight iq2_xxs blk.61.attn_v.weight q4_K blk.62.attn_output.weight iq2_xxs blk.62.attn_v.weight q4_K -blk.63.attn_v.weight q4_K blk.63.attn_output.weight iq2_xxs +blk.63.attn_v.weight q4_K blk.64.attn_output.weight iq2_xxs blk.64.attn_v.weight q4_K blk.65.attn_output.weight iq2_xxs @@ -3313,54 +3314,54 @@ blk.75.attn_output.weight iq2_xxs blk.75.attn_v.weight q4_K blk.76.attn_output.weight iq2_xxs blk.76.attn_v.weight q4_K -blk.77.attn_v.weight q4_K blk.77.attn_output.weight iq2_xxs +blk.77.attn_v.weight q4_K blk.78.attn_output.weight iq2_xxs blk.78.attn_v.weight q4_K blk.79.attn_output.weight iq2_xxs blk.79.attn_v.weight q4_K -output.weight q5_K [BF16] bf16 [TQ1_0] tq1_0 -token_embd.weight q4_K output.weight q6_K +token_embd.weight q4_K [TQ2_0] tq2_0 -token_embd.weight q4_K output.weight q6_K +token_embd.weight q4_K [MXFP4_MOE] mxfp4 +output.weight q8_0 token_embd.weight q8_0 -blk.0.ffn_down.weight q8_0 -blk.0.ffn_gate.weight q8_0 -blk.0.ffn_up.weight q8_0 blk.0.attn_k.weight q8_0 blk.0.attn_output.weight q8_0 blk.0.attn_q.weight q8_0 blk.0.attn_v.weight q8_0 -blk.1.ffn_gate.weight q8_0 +blk.0.ffn_down.weight q8_0 +blk.0.ffn_gate.weight q8_0 +blk.0.ffn_up.weight q8_0 blk.1.attn_k.weight q8_0 blk.1.attn_output.weight q8_0 blk.1.attn_q.weight q8_0 blk.1.attn_v.weight q8_0 blk.1.ffn_down.weight q8_0 +blk.1.ffn_gate.weight q8_0 blk.1.ffn_up.weight q8_0 -blk.2.ffn_down.weight q8_0 -blk.2.ffn_gate.weight q8_0 -blk.2.ffn_up.weight q8_0 blk.2.attn_k.weight q8_0 blk.2.attn_output.weight q8_0 blk.2.attn_q.weight q8_0 blk.2.attn_v.weight q8_0 -blk.3.ffn_down.weight q8_0 -blk.3.ffn_gate.weight q8_0 -blk.3.ffn_up.weight q8_0 +blk.2.ffn_down.weight q8_0 +blk.2.ffn_gate.weight q8_0 +blk.2.ffn_up.weight q8_0 blk.3.attn_k.weight q8_0 blk.3.attn_output.weight q8_0 blk.3.attn_q.weight q8_0 blk.3.attn_v.weight q8_0 +blk.3.ffn_down.weight q8_0 +blk.3.ffn_gate.weight q8_0 +blk.3.ffn_up.weight q8_0 blk.4.attn_k.weight q8_0 blk.4.attn_output.weight q8_0 blk.4.attn_q.weight q8_0 @@ -3368,97 +3369,97 @@ blk.4.attn_v.weight q8_0 blk.4.ffn_down.weight q8_0 blk.4.ffn_gate.weight q8_0 blk.4.ffn_up.weight q8_0 -blk.5.ffn_down.weight q8_0 -blk.5.ffn_gate.weight q8_0 -blk.5.ffn_up.weight q8_0 blk.5.attn_k.weight q8_0 blk.5.attn_output.weight q8_0 blk.5.attn_q.weight q8_0 blk.5.attn_v.weight q8_0 -blk.6.ffn_down.weight q8_0 -blk.6.ffn_gate.weight q8_0 -blk.6.ffn_up.weight q8_0 +blk.5.ffn_down.weight q8_0 +blk.5.ffn_gate.weight q8_0 +blk.5.ffn_up.weight q8_0 blk.6.attn_k.weight q8_0 blk.6.attn_output.weight q8_0 blk.6.attn_q.weight q8_0 blk.6.attn_v.weight q8_0 +blk.6.ffn_down.weight q8_0 +blk.6.ffn_gate.weight q8_0 +blk.6.ffn_up.weight q8_0 blk.7.attn_k.weight q8_0 +blk.7.attn_output.weight q8_0 blk.7.attn_q.weight q8_0 blk.7.attn_v.weight q8_0 blk.7.ffn_down.weight q8_0 blk.7.ffn_gate.weight q8_0 blk.7.ffn_up.weight q8_0 -blk.7.attn_output.weight q8_0 -blk.8.ffn_down.weight q8_0 -blk.8.ffn_gate.weight q8_0 -blk.8.ffn_up.weight q8_0 blk.8.attn_k.weight q8_0 blk.8.attn_output.weight q8_0 blk.8.attn_q.weight q8_0 blk.8.attn_v.weight q8_0 -blk.9.ffn_down.weight q8_0 -blk.9.ffn_gate.weight q8_0 -blk.9.ffn_up.weight q8_0 +blk.8.ffn_down.weight q8_0 +blk.8.ffn_gate.weight q8_0 +blk.8.ffn_up.weight q8_0 blk.9.attn_k.weight q8_0 blk.9.attn_output.weight q8_0 blk.9.attn_q.weight q8_0 blk.9.attn_v.weight q8_0 -blk.10.ffn_down.weight q8_0 -blk.10.ffn_gate.weight q8_0 -blk.10.ffn_up.weight q8_0 +blk.9.ffn_down.weight q8_0 +blk.9.ffn_gate.weight q8_0 +blk.9.ffn_up.weight q8_0 blk.10.attn_k.weight q8_0 blk.10.attn_output.weight q8_0 blk.10.attn_q.weight q8_0 blk.10.attn_v.weight q8_0 -blk.11.ffn_down.weight q8_0 -blk.11.ffn_gate.weight q8_0 -blk.11.ffn_up.weight q8_0 +blk.10.ffn_down.weight q8_0 +blk.10.ffn_gate.weight q8_0 +blk.10.ffn_up.weight q8_0 blk.11.attn_k.weight q8_0 blk.11.attn_output.weight q8_0 blk.11.attn_q.weight q8_0 blk.11.attn_v.weight q8_0 -blk.12.ffn_gate.weight q8_0 -blk.12.ffn_up.weight q8_0 +blk.11.ffn_down.weight q8_0 +blk.11.ffn_gate.weight q8_0 +blk.11.ffn_up.weight q8_0 blk.12.attn_k.weight q8_0 blk.12.attn_output.weight q8_0 blk.12.attn_q.weight q8_0 blk.12.attn_v.weight q8_0 blk.12.ffn_down.weight q8_0 -blk.13.ffn_down.weight q8_0 -blk.13.ffn_gate.weight q8_0 -blk.13.ffn_up.weight q8_0 +blk.12.ffn_gate.weight q8_0 +blk.12.ffn_up.weight q8_0 blk.13.attn_k.weight q8_0 blk.13.attn_output.weight q8_0 blk.13.attn_q.weight q8_0 blk.13.attn_v.weight q8_0 -blk.14.ffn_down.weight q8_0 -blk.14.ffn_gate.weight q8_0 -blk.14.ffn_up.weight q8_0 +blk.13.ffn_down.weight q8_0 +blk.13.ffn_gate.weight q8_0 +blk.13.ffn_up.weight q8_0 blk.14.attn_k.weight q8_0 blk.14.attn_output.weight q8_0 blk.14.attn_q.weight q8_0 blk.14.attn_v.weight q8_0 -blk.15.ffn_gate.weight q8_0 +blk.14.ffn_down.weight q8_0 +blk.14.ffn_gate.weight q8_0 +blk.14.ffn_up.weight q8_0 blk.15.attn_k.weight q8_0 blk.15.attn_output.weight q8_0 blk.15.attn_q.weight q8_0 blk.15.attn_v.weight q8_0 blk.15.ffn_down.weight q8_0 +blk.15.ffn_gate.weight q8_0 blk.15.ffn_up.weight q8_0 -blk.16.ffn_down.weight q8_0 -blk.16.ffn_gate.weight q8_0 -blk.16.ffn_up.weight q8_0 blk.16.attn_k.weight q8_0 blk.16.attn_output.weight q8_0 blk.16.attn_q.weight q8_0 blk.16.attn_v.weight q8_0 -blk.17.ffn_down.weight q8_0 -blk.17.ffn_gate.weight q8_0 -blk.17.ffn_up.weight q8_0 +blk.16.ffn_down.weight q8_0 +blk.16.ffn_gate.weight q8_0 +blk.16.ffn_up.weight q8_0 blk.17.attn_k.weight q8_0 blk.17.attn_output.weight q8_0 blk.17.attn_q.weight q8_0 blk.17.attn_v.weight q8_0 +blk.17.ffn_down.weight q8_0 +blk.17.ffn_gate.weight q8_0 +blk.17.ffn_up.weight q8_0 blk.18.attn_k.weight q8_0 blk.18.attn_output.weight q8_0 blk.18.attn_q.weight q8_0 @@ -3466,97 +3467,97 @@ blk.18.attn_v.weight q8_0 blk.18.ffn_down.weight q8_0 blk.18.ffn_gate.weight q8_0 blk.18.ffn_up.weight q8_0 -blk.19.ffn_down.weight q8_0 -blk.19.ffn_gate.weight q8_0 -blk.19.ffn_up.weight q8_0 blk.19.attn_k.weight q8_0 blk.19.attn_output.weight q8_0 blk.19.attn_q.weight q8_0 blk.19.attn_v.weight q8_0 -blk.20.ffn_down.weight q8_0 -blk.20.ffn_gate.weight q8_0 -blk.20.ffn_up.weight q8_0 +blk.19.ffn_down.weight q8_0 +blk.19.ffn_gate.weight q8_0 +blk.19.ffn_up.weight q8_0 blk.20.attn_k.weight q8_0 blk.20.attn_output.weight q8_0 blk.20.attn_q.weight q8_0 blk.20.attn_v.weight q8_0 +blk.20.ffn_down.weight q8_0 +blk.20.ffn_gate.weight q8_0 +blk.20.ffn_up.weight q8_0 blk.21.attn_k.weight q8_0 +blk.21.attn_output.weight q8_0 blk.21.attn_q.weight q8_0 blk.21.attn_v.weight q8_0 blk.21.ffn_down.weight q8_0 blk.21.ffn_gate.weight q8_0 blk.21.ffn_up.weight q8_0 -blk.21.attn_output.weight q8_0 -blk.22.ffn_down.weight q8_0 -blk.22.ffn_gate.weight q8_0 -blk.22.ffn_up.weight q8_0 blk.22.attn_k.weight q8_0 blk.22.attn_output.weight q8_0 blk.22.attn_q.weight q8_0 blk.22.attn_v.weight q8_0 -blk.23.ffn_down.weight q8_0 -blk.23.ffn_gate.weight q8_0 -blk.23.ffn_up.weight q8_0 +blk.22.ffn_down.weight q8_0 +blk.22.ffn_gate.weight q8_0 +blk.22.ffn_up.weight q8_0 blk.23.attn_k.weight q8_0 blk.23.attn_output.weight q8_0 blk.23.attn_q.weight q8_0 blk.23.attn_v.weight q8_0 -blk.24.ffn_down.weight q8_0 -blk.24.ffn_gate.weight q8_0 -blk.24.ffn_up.weight q8_0 +blk.23.ffn_down.weight q8_0 +blk.23.ffn_gate.weight q8_0 +blk.23.ffn_up.weight q8_0 blk.24.attn_k.weight q8_0 blk.24.attn_output.weight q8_0 blk.24.attn_q.weight q8_0 blk.24.attn_v.weight q8_0 -blk.25.ffn_down.weight q8_0 -blk.25.ffn_gate.weight q8_0 -blk.25.ffn_up.weight q8_0 +blk.24.ffn_down.weight q8_0 +blk.24.ffn_gate.weight q8_0 +blk.24.ffn_up.weight q8_0 blk.25.attn_k.weight q8_0 blk.25.attn_output.weight q8_0 blk.25.attn_q.weight q8_0 blk.25.attn_v.weight q8_0 -blk.26.ffn_gate.weight q8_0 -blk.26.ffn_up.weight q8_0 +blk.25.ffn_down.weight q8_0 +blk.25.ffn_gate.weight q8_0 +blk.25.ffn_up.weight q8_0 blk.26.attn_k.weight q8_0 blk.26.attn_output.weight q8_0 blk.26.attn_q.weight q8_0 blk.26.attn_v.weight q8_0 blk.26.ffn_down.weight q8_0 -blk.27.ffn_down.weight q8_0 -blk.27.ffn_gate.weight q8_0 -blk.27.ffn_up.weight q8_0 +blk.26.ffn_gate.weight q8_0 +blk.26.ffn_up.weight q8_0 blk.27.attn_k.weight q8_0 blk.27.attn_output.weight q8_0 blk.27.attn_q.weight q8_0 blk.27.attn_v.weight q8_0 -blk.28.ffn_down.weight q8_0 -blk.28.ffn_gate.weight q8_0 -blk.28.ffn_up.weight q8_0 +blk.27.ffn_down.weight q8_0 +blk.27.ffn_gate.weight q8_0 +blk.27.ffn_up.weight q8_0 blk.28.attn_k.weight q8_0 blk.28.attn_output.weight q8_0 blk.28.attn_q.weight q8_0 blk.28.attn_v.weight q8_0 -blk.29.ffn_gate.weight q8_0 +blk.28.ffn_down.weight q8_0 +blk.28.ffn_gate.weight q8_0 +blk.28.ffn_up.weight q8_0 blk.29.attn_k.weight q8_0 blk.29.attn_output.weight q8_0 blk.29.attn_q.weight q8_0 blk.29.attn_v.weight q8_0 blk.29.ffn_down.weight q8_0 +blk.29.ffn_gate.weight q8_0 blk.29.ffn_up.weight q8_0 -blk.30.ffn_down.weight q8_0 -blk.30.ffn_gate.weight q8_0 -blk.30.ffn_up.weight q8_0 blk.30.attn_k.weight q8_0 blk.30.attn_output.weight q8_0 blk.30.attn_q.weight q8_0 blk.30.attn_v.weight q8_0 -blk.31.ffn_down.weight q8_0 -blk.31.ffn_gate.weight q8_0 -blk.31.ffn_up.weight q8_0 +blk.30.ffn_down.weight q8_0 +blk.30.ffn_gate.weight q8_0 +blk.30.ffn_up.weight q8_0 blk.31.attn_k.weight q8_0 blk.31.attn_output.weight q8_0 blk.31.attn_q.weight q8_0 blk.31.attn_v.weight q8_0 +blk.31.ffn_down.weight q8_0 +blk.31.ffn_gate.weight q8_0 +blk.31.ffn_up.weight q8_0 blk.32.attn_k.weight q8_0 blk.32.attn_output.weight q8_0 blk.32.attn_q.weight q8_0 @@ -3564,97 +3565,97 @@ blk.32.attn_v.weight q8_0 blk.32.ffn_down.weight q8_0 blk.32.ffn_gate.weight q8_0 blk.32.ffn_up.weight q8_0 -blk.33.ffn_down.weight q8_0 -blk.33.ffn_gate.weight q8_0 -blk.33.ffn_up.weight q8_0 blk.33.attn_k.weight q8_0 blk.33.attn_output.weight q8_0 blk.33.attn_q.weight q8_0 blk.33.attn_v.weight q8_0 -blk.34.ffn_down.weight q8_0 -blk.34.ffn_gate.weight q8_0 -blk.34.ffn_up.weight q8_0 +blk.33.ffn_down.weight q8_0 +blk.33.ffn_gate.weight q8_0 +blk.33.ffn_up.weight q8_0 blk.34.attn_k.weight q8_0 blk.34.attn_output.weight q8_0 blk.34.attn_q.weight q8_0 blk.34.attn_v.weight q8_0 +blk.34.ffn_down.weight q8_0 +blk.34.ffn_gate.weight q8_0 +blk.34.ffn_up.weight q8_0 blk.35.attn_k.weight q8_0 +blk.35.attn_output.weight q8_0 blk.35.attn_q.weight q8_0 blk.35.attn_v.weight q8_0 blk.35.ffn_down.weight q8_0 blk.35.ffn_gate.weight q8_0 blk.35.ffn_up.weight q8_0 -blk.35.attn_output.weight q8_0 -blk.36.ffn_down.weight q8_0 -blk.36.ffn_gate.weight q8_0 -blk.36.ffn_up.weight q8_0 blk.36.attn_k.weight q8_0 blk.36.attn_output.weight q8_0 blk.36.attn_q.weight q8_0 blk.36.attn_v.weight q8_0 -blk.37.ffn_down.weight q8_0 -blk.37.ffn_gate.weight q8_0 -blk.37.ffn_up.weight q8_0 +blk.36.ffn_down.weight q8_0 +blk.36.ffn_gate.weight q8_0 +blk.36.ffn_up.weight q8_0 blk.37.attn_k.weight q8_0 blk.37.attn_output.weight q8_0 blk.37.attn_q.weight q8_0 blk.37.attn_v.weight q8_0 -blk.38.ffn_down.weight q8_0 -blk.38.ffn_gate.weight q8_0 -blk.38.ffn_up.weight q8_0 +blk.37.ffn_down.weight q8_0 +blk.37.ffn_gate.weight q8_0 +blk.37.ffn_up.weight q8_0 blk.38.attn_k.weight q8_0 blk.38.attn_output.weight q8_0 blk.38.attn_q.weight q8_0 blk.38.attn_v.weight q8_0 -blk.39.ffn_down.weight q8_0 -blk.39.ffn_gate.weight q8_0 -blk.39.ffn_up.weight q8_0 +blk.38.ffn_down.weight q8_0 +blk.38.ffn_gate.weight q8_0 +blk.38.ffn_up.weight q8_0 blk.39.attn_k.weight q8_0 blk.39.attn_output.weight q8_0 blk.39.attn_q.weight q8_0 blk.39.attn_v.weight q8_0 -blk.40.ffn_gate.weight q8_0 -blk.40.ffn_up.weight q8_0 +blk.39.ffn_down.weight q8_0 +blk.39.ffn_gate.weight q8_0 +blk.39.ffn_up.weight q8_0 blk.40.attn_k.weight q8_0 blk.40.attn_output.weight q8_0 blk.40.attn_q.weight q8_0 blk.40.attn_v.weight q8_0 blk.40.ffn_down.weight q8_0 -blk.41.ffn_down.weight q8_0 -blk.41.ffn_gate.weight q8_0 -blk.41.ffn_up.weight q8_0 +blk.40.ffn_gate.weight q8_0 +blk.40.ffn_up.weight q8_0 blk.41.attn_k.weight q8_0 blk.41.attn_output.weight q8_0 blk.41.attn_q.weight q8_0 blk.41.attn_v.weight q8_0 -blk.42.ffn_down.weight q8_0 -blk.42.ffn_gate.weight q8_0 -blk.42.ffn_up.weight q8_0 +blk.41.ffn_down.weight q8_0 +blk.41.ffn_gate.weight q8_0 +blk.41.ffn_up.weight q8_0 blk.42.attn_k.weight q8_0 blk.42.attn_output.weight q8_0 blk.42.attn_q.weight q8_0 blk.42.attn_v.weight q8_0 -blk.43.ffn_gate.weight q8_0 +blk.42.ffn_down.weight q8_0 +blk.42.ffn_gate.weight q8_0 +blk.42.ffn_up.weight q8_0 blk.43.attn_k.weight q8_0 blk.43.attn_output.weight q8_0 blk.43.attn_q.weight q8_0 blk.43.attn_v.weight q8_0 blk.43.ffn_down.weight q8_0 +blk.43.ffn_gate.weight q8_0 blk.43.ffn_up.weight q8_0 -blk.44.ffn_down.weight q8_0 -blk.44.ffn_gate.weight q8_0 -blk.44.ffn_up.weight q8_0 blk.44.attn_k.weight q8_0 blk.44.attn_output.weight q8_0 blk.44.attn_q.weight q8_0 blk.44.attn_v.weight q8_0 -blk.45.ffn_down.weight q8_0 -blk.45.ffn_gate.weight q8_0 -blk.45.ffn_up.weight q8_0 +blk.44.ffn_down.weight q8_0 +blk.44.ffn_gate.weight q8_0 +blk.44.ffn_up.weight q8_0 blk.45.attn_k.weight q8_0 blk.45.attn_output.weight q8_0 blk.45.attn_q.weight q8_0 blk.45.attn_v.weight q8_0 +blk.45.ffn_down.weight q8_0 +blk.45.ffn_gate.weight q8_0 +blk.45.ffn_up.weight q8_0 blk.46.attn_k.weight q8_0 blk.46.attn_output.weight q8_0 blk.46.attn_q.weight q8_0 @@ -3662,97 +3663,97 @@ blk.46.attn_v.weight q8_0 blk.46.ffn_down.weight q8_0 blk.46.ffn_gate.weight q8_0 blk.46.ffn_up.weight q8_0 -blk.47.ffn_down.weight q8_0 -blk.47.ffn_gate.weight q8_0 -blk.47.ffn_up.weight q8_0 blk.47.attn_k.weight q8_0 blk.47.attn_output.weight q8_0 blk.47.attn_q.weight q8_0 blk.47.attn_v.weight q8_0 -blk.48.ffn_down.weight q8_0 -blk.48.ffn_gate.weight q8_0 -blk.48.ffn_up.weight q8_0 +blk.47.ffn_down.weight q8_0 +blk.47.ffn_gate.weight q8_0 +blk.47.ffn_up.weight q8_0 blk.48.attn_k.weight q8_0 blk.48.attn_output.weight q8_0 blk.48.attn_q.weight q8_0 blk.48.attn_v.weight q8_0 +blk.48.ffn_down.weight q8_0 +blk.48.ffn_gate.weight q8_0 +blk.48.ffn_up.weight q8_0 blk.49.attn_k.weight q8_0 +blk.49.attn_output.weight q8_0 blk.49.attn_q.weight q8_0 blk.49.attn_v.weight q8_0 blk.49.ffn_down.weight q8_0 blk.49.ffn_gate.weight q8_0 blk.49.ffn_up.weight q8_0 -blk.49.attn_output.weight q8_0 -blk.50.ffn_down.weight q8_0 -blk.50.ffn_gate.weight q8_0 -blk.50.ffn_up.weight q8_0 blk.50.attn_k.weight q8_0 blk.50.attn_output.weight q8_0 blk.50.attn_q.weight q8_0 blk.50.attn_v.weight q8_0 -blk.51.ffn_down.weight q8_0 -blk.51.ffn_gate.weight q8_0 -blk.51.ffn_up.weight q8_0 +blk.50.ffn_down.weight q8_0 +blk.50.ffn_gate.weight q8_0 +blk.50.ffn_up.weight q8_0 blk.51.attn_k.weight q8_0 blk.51.attn_output.weight q8_0 blk.51.attn_q.weight q8_0 blk.51.attn_v.weight q8_0 -blk.52.ffn_down.weight q8_0 -blk.52.ffn_gate.weight q8_0 -blk.52.ffn_up.weight q8_0 +blk.51.ffn_down.weight q8_0 +blk.51.ffn_gate.weight q8_0 +blk.51.ffn_up.weight q8_0 blk.52.attn_k.weight q8_0 blk.52.attn_output.weight q8_0 blk.52.attn_q.weight q8_0 blk.52.attn_v.weight q8_0 -blk.53.ffn_down.weight q8_0 -blk.53.ffn_gate.weight q8_0 -blk.53.ffn_up.weight q8_0 +blk.52.ffn_down.weight q8_0 +blk.52.ffn_gate.weight q8_0 +blk.52.ffn_up.weight q8_0 blk.53.attn_k.weight q8_0 blk.53.attn_output.weight q8_0 blk.53.attn_q.weight q8_0 blk.53.attn_v.weight q8_0 -blk.54.ffn_gate.weight q8_0 -blk.54.ffn_up.weight q8_0 +blk.53.ffn_down.weight q8_0 +blk.53.ffn_gate.weight q8_0 +blk.53.ffn_up.weight q8_0 blk.54.attn_k.weight q8_0 blk.54.attn_output.weight q8_0 blk.54.attn_q.weight q8_0 blk.54.attn_v.weight q8_0 blk.54.ffn_down.weight q8_0 -blk.55.ffn_down.weight q8_0 -blk.55.ffn_gate.weight q8_0 -blk.55.ffn_up.weight q8_0 +blk.54.ffn_gate.weight q8_0 +blk.54.ffn_up.weight q8_0 blk.55.attn_k.weight q8_0 blk.55.attn_output.weight q8_0 blk.55.attn_q.weight q8_0 blk.55.attn_v.weight q8_0 -blk.56.ffn_down.weight q8_0 -blk.56.ffn_gate.weight q8_0 -blk.56.ffn_up.weight q8_0 +blk.55.ffn_down.weight q8_0 +blk.55.ffn_gate.weight q8_0 +blk.55.ffn_up.weight q8_0 blk.56.attn_k.weight q8_0 blk.56.attn_output.weight q8_0 blk.56.attn_q.weight q8_0 blk.56.attn_v.weight q8_0 -blk.57.ffn_gate.weight q8_0 +blk.56.ffn_down.weight q8_0 +blk.56.ffn_gate.weight q8_0 +blk.56.ffn_up.weight q8_0 blk.57.attn_k.weight q8_0 blk.57.attn_output.weight q8_0 blk.57.attn_q.weight q8_0 blk.57.attn_v.weight q8_0 blk.57.ffn_down.weight q8_0 +blk.57.ffn_gate.weight q8_0 blk.57.ffn_up.weight q8_0 -blk.58.ffn_down.weight q8_0 -blk.58.ffn_gate.weight q8_0 -blk.58.ffn_up.weight q8_0 blk.58.attn_k.weight q8_0 blk.58.attn_output.weight q8_0 blk.58.attn_q.weight q8_0 blk.58.attn_v.weight q8_0 -blk.59.ffn_down.weight q8_0 -blk.59.ffn_gate.weight q8_0 -blk.59.ffn_up.weight q8_0 +blk.58.ffn_down.weight q8_0 +blk.58.ffn_gate.weight q8_0 +blk.58.ffn_up.weight q8_0 blk.59.attn_k.weight q8_0 blk.59.attn_output.weight q8_0 blk.59.attn_q.weight q8_0 blk.59.attn_v.weight q8_0 +blk.59.ffn_down.weight q8_0 +blk.59.ffn_gate.weight q8_0 +blk.59.ffn_up.weight q8_0 blk.60.attn_k.weight q8_0 blk.60.attn_output.weight q8_0 blk.60.attn_q.weight q8_0 @@ -3760,97 +3761,97 @@ blk.60.attn_v.weight q8_0 blk.60.ffn_down.weight q8_0 blk.60.ffn_gate.weight q8_0 blk.60.ffn_up.weight q8_0 -blk.61.ffn_down.weight q8_0 -blk.61.ffn_gate.weight q8_0 -blk.61.ffn_up.weight q8_0 blk.61.attn_k.weight q8_0 blk.61.attn_output.weight q8_0 blk.61.attn_q.weight q8_0 blk.61.attn_v.weight q8_0 -blk.62.ffn_down.weight q8_0 -blk.62.ffn_gate.weight q8_0 -blk.62.ffn_up.weight q8_0 +blk.61.ffn_down.weight q8_0 +blk.61.ffn_gate.weight q8_0 +blk.61.ffn_up.weight q8_0 blk.62.attn_k.weight q8_0 blk.62.attn_output.weight q8_0 blk.62.attn_q.weight q8_0 blk.62.attn_v.weight q8_0 +blk.62.ffn_down.weight q8_0 +blk.62.ffn_gate.weight q8_0 +blk.62.ffn_up.weight q8_0 blk.63.attn_k.weight q8_0 +blk.63.attn_output.weight q8_0 blk.63.attn_q.weight q8_0 blk.63.attn_v.weight q8_0 blk.63.ffn_down.weight q8_0 blk.63.ffn_gate.weight q8_0 blk.63.ffn_up.weight q8_0 -blk.63.attn_output.weight q8_0 -blk.64.ffn_down.weight q8_0 -blk.64.ffn_gate.weight q8_0 -blk.64.ffn_up.weight q8_0 blk.64.attn_k.weight q8_0 blk.64.attn_output.weight q8_0 blk.64.attn_q.weight q8_0 blk.64.attn_v.weight q8_0 -blk.65.ffn_down.weight q8_0 -blk.65.ffn_gate.weight q8_0 -blk.65.ffn_up.weight q8_0 +blk.64.ffn_down.weight q8_0 +blk.64.ffn_gate.weight q8_0 +blk.64.ffn_up.weight q8_0 blk.65.attn_k.weight q8_0 blk.65.attn_output.weight q8_0 blk.65.attn_q.weight q8_0 blk.65.attn_v.weight q8_0 -blk.66.ffn_down.weight q8_0 -blk.66.ffn_gate.weight q8_0 -blk.66.ffn_up.weight q8_0 +blk.65.ffn_down.weight q8_0 +blk.65.ffn_gate.weight q8_0 +blk.65.ffn_up.weight q8_0 blk.66.attn_k.weight q8_0 blk.66.attn_output.weight q8_0 blk.66.attn_q.weight q8_0 blk.66.attn_v.weight q8_0 -blk.67.ffn_down.weight q8_0 -blk.67.ffn_gate.weight q8_0 -blk.67.ffn_up.weight q8_0 +blk.66.ffn_down.weight q8_0 +blk.66.ffn_gate.weight q8_0 +blk.66.ffn_up.weight q8_0 blk.67.attn_k.weight q8_0 blk.67.attn_output.weight q8_0 blk.67.attn_q.weight q8_0 blk.67.attn_v.weight q8_0 -blk.68.ffn_gate.weight q8_0 -blk.68.ffn_up.weight q8_0 +blk.67.ffn_down.weight q8_0 +blk.67.ffn_gate.weight q8_0 +blk.67.ffn_up.weight q8_0 blk.68.attn_k.weight q8_0 blk.68.attn_output.weight q8_0 blk.68.attn_q.weight q8_0 blk.68.attn_v.weight q8_0 blk.68.ffn_down.weight q8_0 -blk.69.ffn_down.weight q8_0 -blk.69.ffn_gate.weight q8_0 -blk.69.ffn_up.weight q8_0 +blk.68.ffn_gate.weight q8_0 +blk.68.ffn_up.weight q8_0 blk.69.attn_k.weight q8_0 blk.69.attn_output.weight q8_0 blk.69.attn_q.weight q8_0 blk.69.attn_v.weight q8_0 -blk.70.ffn_down.weight q8_0 -blk.70.ffn_gate.weight q8_0 -blk.70.ffn_up.weight q8_0 +blk.69.ffn_down.weight q8_0 +blk.69.ffn_gate.weight q8_0 +blk.69.ffn_up.weight q8_0 blk.70.attn_k.weight q8_0 blk.70.attn_output.weight q8_0 blk.70.attn_q.weight q8_0 blk.70.attn_v.weight q8_0 -blk.71.ffn_gate.weight q8_0 +blk.70.ffn_down.weight q8_0 +blk.70.ffn_gate.weight q8_0 +blk.70.ffn_up.weight q8_0 blk.71.attn_k.weight q8_0 blk.71.attn_output.weight q8_0 blk.71.attn_q.weight q8_0 blk.71.attn_v.weight q8_0 blk.71.ffn_down.weight q8_0 +blk.71.ffn_gate.weight q8_0 blk.71.ffn_up.weight q8_0 -blk.72.ffn_down.weight q8_0 -blk.72.ffn_gate.weight q8_0 -blk.72.ffn_up.weight q8_0 blk.72.attn_k.weight q8_0 blk.72.attn_output.weight q8_0 blk.72.attn_q.weight q8_0 blk.72.attn_v.weight q8_0 -blk.73.ffn_down.weight q8_0 -blk.73.ffn_gate.weight q8_0 -blk.73.ffn_up.weight q8_0 +blk.72.ffn_down.weight q8_0 +blk.72.ffn_gate.weight q8_0 +blk.72.ffn_up.weight q8_0 blk.73.attn_k.weight q8_0 blk.73.attn_output.weight q8_0 blk.73.attn_q.weight q8_0 blk.73.attn_v.weight q8_0 +blk.73.ffn_down.weight q8_0 +blk.73.ffn_gate.weight q8_0 +blk.73.ffn_up.weight q8_0 blk.74.attn_k.weight q8_0 blk.74.attn_output.weight q8_0 blk.74.attn_q.weight q8_0 @@ -3858,39 +3859,38 @@ blk.74.attn_v.weight q8_0 blk.74.ffn_down.weight q8_0 blk.74.ffn_gate.weight q8_0 blk.74.ffn_up.weight q8_0 -blk.75.ffn_down.weight q8_0 -blk.75.ffn_gate.weight q8_0 -blk.75.ffn_up.weight q8_0 blk.75.attn_k.weight q8_0 blk.75.attn_output.weight q8_0 blk.75.attn_q.weight q8_0 blk.75.attn_v.weight q8_0 -blk.76.ffn_down.weight q8_0 -blk.76.ffn_gate.weight q8_0 -blk.76.ffn_up.weight q8_0 +blk.75.ffn_down.weight q8_0 +blk.75.ffn_gate.weight q8_0 +blk.75.ffn_up.weight q8_0 blk.76.attn_k.weight q8_0 blk.76.attn_output.weight q8_0 blk.76.attn_q.weight q8_0 blk.76.attn_v.weight q8_0 +blk.76.ffn_down.weight q8_0 +blk.76.ffn_gate.weight q8_0 +blk.76.ffn_up.weight q8_0 blk.77.attn_k.weight q8_0 +blk.77.attn_output.weight q8_0 blk.77.attn_q.weight q8_0 blk.77.attn_v.weight q8_0 blk.77.ffn_down.weight q8_0 blk.77.ffn_gate.weight q8_0 blk.77.ffn_up.weight q8_0 -blk.77.attn_output.weight q8_0 -blk.78.ffn_down.weight q8_0 -blk.78.ffn_gate.weight q8_0 -blk.78.ffn_up.weight q8_0 blk.78.attn_k.weight q8_0 blk.78.attn_output.weight q8_0 blk.78.attn_q.weight q8_0 blk.78.attn_v.weight q8_0 -blk.79.ffn_down.weight q8_0 -blk.79.ffn_gate.weight q8_0 -blk.79.ffn_up.weight q8_0 +blk.78.ffn_down.weight q8_0 +blk.78.ffn_gate.weight q8_0 +blk.78.ffn_up.weight q8_0 blk.79.attn_k.weight q8_0 blk.79.attn_output.weight q8_0 blk.79.attn_q.weight q8_0 blk.79.attn_v.weight q8_0 -output.weight q8_0 +blk.79.ffn_down.weight q8_0 +blk.79.ffn_gate.weight q8_0 +blk.79.ffn_up.weight q8_0 diff --git a/tests/snapshots/qwen3-coder-next.schema b/tests/snapshots/qwen3-coder-next.schema index 9bf16fca4b..3862a48ce6 100644 --- a/tests/snapshots/qwen3-coder-next.schema +++ b/tests/snapshots/qwen3-coder-next.schema @@ -20,670 +20,671 @@ output.weight q6_K output.weight q6_K [Q2_K] q2_K +output.weight q6_K blk.0.attn_qkv.weight q4_K blk.0.ffn_down_exps.weight q3_K blk.0.ffn_down_shexp.weight q3_K blk.1.attn_qkv.weight q4_K -blk.1.ffn_down_shexp.weight q3_K blk.1.ffn_down_exps.weight q3_K +blk.1.ffn_down_shexp.weight q3_K blk.2.attn_qkv.weight q4_K -blk.2.ffn_down_shexp.weight q3_K blk.2.ffn_down_exps.weight q3_K -blk.3.ffn_down_shexp.weight q3_K +blk.2.ffn_down_shexp.weight q3_K blk.3.attn_output.weight q3_K blk.3.attn_v.weight q4_K blk.3.ffn_down_exps.weight q3_K +blk.3.ffn_down_shexp.weight q3_K blk.4.attn_qkv.weight q4_K -blk.4.ffn_down_shexp.weight q3_K blk.4.ffn_down_exps.weight q3_K +blk.4.ffn_down_shexp.weight q3_K blk.5.attn_qkv.weight q4_K -blk.5.ffn_down_shexp.weight q3_K blk.5.ffn_down_exps.weight q3_K +blk.5.ffn_down_shexp.weight q3_K blk.6.attn_qkv.weight q4_K blk.6.ffn_down_exps.weight q3_K blk.6.ffn_down_shexp.weight q3_K -blk.7.ffn_down_shexp.weight q3_K blk.7.attn_output.weight q3_K blk.7.attn_v.weight q4_K blk.7.ffn_down_exps.weight q3_K +blk.7.ffn_down_shexp.weight q3_K blk.8.attn_qkv.weight q4_K -blk.8.ffn_down_shexp.weight q3_K blk.8.ffn_down_exps.weight q3_K +blk.8.ffn_down_shexp.weight q3_K blk.9.attn_qkv.weight q4_K +blk.9.ffn_down_exps.weight q3_K blk.9.ffn_down_shexp.weight q3_K blk.10.attn_qkv.weight q4_K -blk.10.ffn_down_shexp.weight q3_K -blk.9.ffn_down_exps.weight q3_K blk.10.ffn_down_exps.weight q3_K -blk.11.ffn_down_shexp.weight q3_K +blk.10.ffn_down_shexp.weight q3_K blk.11.attn_output.weight q3_K blk.11.attn_v.weight q4_K blk.11.ffn_down_exps.weight q3_K +blk.11.ffn_down_shexp.weight q3_K blk.12.attn_qkv.weight q4_K blk.12.ffn_down_exps.weight q3_K blk.12.ffn_down_shexp.weight q3_K blk.13.attn_qkv.weight q4_K -blk.13.ffn_down_shexp.weight q3_K blk.13.ffn_down_exps.weight q3_K +blk.13.ffn_down_shexp.weight q3_K blk.14.attn_qkv.weight q4_K -blk.14.ffn_down_shexp.weight q3_K blk.14.ffn_down_exps.weight q3_K -blk.15.ffn_down_shexp.weight q3_K +blk.14.ffn_down_shexp.weight q3_K blk.15.attn_output.weight q3_K blk.15.attn_v.weight q4_K blk.15.ffn_down_exps.weight q3_K +blk.15.ffn_down_shexp.weight q3_K blk.16.attn_qkv.weight q4_K -blk.16.ffn_down_shexp.weight q3_K blk.16.ffn_down_exps.weight q3_K +blk.16.ffn_down_shexp.weight q3_K blk.17.attn_qkv.weight q4_K blk.17.ffn_down_exps.weight q3_K blk.17.ffn_down_shexp.weight q3_K blk.18.attn_qkv.weight q4_K -blk.18.ffn_down_shexp.weight q3_K blk.18.ffn_down_exps.weight q3_K -blk.19.ffn_down_shexp.weight q3_K +blk.18.ffn_down_shexp.weight q3_K blk.19.attn_output.weight q3_K blk.19.attn_v.weight q4_K blk.19.ffn_down_exps.weight q3_K +blk.19.ffn_down_shexp.weight q3_K blk.20.attn_qkv.weight q4_K -blk.20.ffn_down_shexp.weight q3_K blk.20.ffn_down_exps.weight q3_K +blk.20.ffn_down_shexp.weight q3_K blk.21.attn_qkv.weight q4_K -blk.21.ffn_down_shexp.weight q3_K blk.21.ffn_down_exps.weight q3_K +blk.21.ffn_down_shexp.weight q3_K blk.22.attn_qkv.weight q4_K -blk.22.ffn_down_shexp.weight q3_K blk.22.ffn_down_exps.weight q3_K -blk.23.ffn_down_exps.weight q3_K -blk.23.ffn_down_shexp.weight q3_K +blk.22.ffn_down_shexp.weight q3_K blk.23.attn_output.weight q3_K blk.23.attn_v.weight q4_K +blk.23.ffn_down_exps.weight q3_K +blk.23.ffn_down_shexp.weight q3_K blk.24.attn_qkv.weight q4_K -blk.24.ffn_down_shexp.weight q3_K blk.24.ffn_down_exps.weight q3_K +blk.24.ffn_down_shexp.weight q3_K blk.25.attn_qkv.weight q4_K -blk.25.ffn_down_shexp.weight q3_K blk.25.ffn_down_exps.weight q3_K +blk.25.ffn_down_shexp.weight q3_K blk.26.attn_qkv.weight q4_K -blk.26.ffn_down_shexp.weight q3_K blk.26.ffn_down_exps.weight q3_K -blk.27.ffn_down_shexp.weight q3_K +blk.26.ffn_down_shexp.weight q3_K blk.27.attn_output.weight q3_K blk.27.attn_v.weight q4_K blk.27.ffn_down_exps.weight q3_K +blk.27.ffn_down_shexp.weight q3_K blk.28.attn_qkv.weight q4_K -blk.28.ffn_down_shexp.weight q3_K blk.28.ffn_down_exps.weight q3_K +blk.28.ffn_down_shexp.weight q3_K blk.29.attn_qkv.weight q4_K blk.29.ffn_down_exps.weight q3_K blk.29.ffn_down_shexp.weight q3_K blk.30.attn_qkv.weight q4_K -blk.30.ffn_down_shexp.weight q3_K blk.30.ffn_down_exps.weight q3_K -blk.31.ffn_down_shexp.weight q3_K +blk.30.ffn_down_shexp.weight q3_K blk.31.attn_output.weight q3_K blk.31.attn_v.weight q4_K blk.31.ffn_down_exps.weight q3_K +blk.31.ffn_down_shexp.weight q3_K blk.32.attn_qkv.weight q4_K -blk.32.ffn_down_shexp.weight q3_K blk.32.ffn_down_exps.weight q3_K +blk.32.ffn_down_shexp.weight q3_K blk.33.attn_qkv.weight q4_K -blk.33.ffn_down_shexp.weight q3_K blk.33.ffn_down_exps.weight q3_K +blk.33.ffn_down_shexp.weight q3_K blk.34.attn_qkv.weight q4_K blk.34.ffn_down_exps.weight q3_K blk.34.ffn_down_shexp.weight q3_K -blk.35.ffn_down_shexp.weight q3_K blk.35.attn_output.weight q3_K blk.35.attn_v.weight q4_K blk.35.ffn_down_exps.weight q3_K +blk.35.ffn_down_shexp.weight q3_K blk.36.attn_qkv.weight q4_K -blk.36.ffn_down_shexp.weight q3_K blk.36.ffn_down_exps.weight q3_K +blk.36.ffn_down_shexp.weight q3_K blk.37.attn_qkv.weight q4_K -blk.37.ffn_down_shexp.weight q3_K blk.37.ffn_down_exps.weight q3_K +blk.37.ffn_down_shexp.weight q3_K blk.38.attn_qkv.weight q4_K -blk.38.ffn_down_shexp.weight q3_K blk.38.ffn_down_exps.weight q3_K -blk.39.ffn_down_shexp.weight q3_K +blk.38.ffn_down_shexp.weight q3_K blk.39.attn_output.weight q3_K blk.39.attn_v.weight q4_K blk.39.ffn_down_exps.weight q3_K +blk.39.ffn_down_shexp.weight q3_K blk.40.attn_qkv.weight q4_K blk.40.ffn_down_exps.weight q3_K blk.40.ffn_down_shexp.weight q3_K blk.41.attn_qkv.weight q4_K -blk.41.ffn_down_shexp.weight q3_K blk.41.ffn_down_exps.weight q3_K +blk.41.ffn_down_shexp.weight q3_K blk.42.attn_qkv.weight q4_K -blk.42.ffn_down_shexp.weight q3_K blk.42.ffn_down_exps.weight q3_K -blk.43.ffn_down_shexp.weight q3_K +blk.42.ffn_down_shexp.weight q3_K blk.43.attn_output.weight q3_K blk.43.attn_v.weight q4_K blk.43.ffn_down_exps.weight q3_K +blk.43.ffn_down_shexp.weight q3_K blk.44.attn_qkv.weight q4_K -blk.44.ffn_down_shexp.weight q3_K blk.44.ffn_down_exps.weight q3_K +blk.44.ffn_down_shexp.weight q3_K blk.45.attn_qkv.weight q4_K -blk.45.ffn_down_shexp.weight q3_K blk.45.ffn_down_exps.weight q3_K +blk.45.ffn_down_shexp.weight q3_K blk.46.attn_qkv.weight q4_K blk.46.ffn_down_exps.weight q3_K blk.46.ffn_down_shexp.weight q3_K -blk.47.ffn_down_shexp.weight q3_K blk.47.attn_output.weight q3_K blk.47.attn_v.weight q4_K -output.weight q6_K blk.47.ffn_down_exps.weight q3_K +blk.47.ffn_down_shexp.weight q3_K [Q3_K_S] q3_K output.weight q6_K [Q3_K_M] q3_K +output.weight q6_K blk.0.attn_qkv.weight q5_K blk.0.ffn_down_exps.weight q5_K blk.0.ffn_down_shexp.weight q5_K blk.1.attn_qkv.weight q5_K -blk.1.ffn_down_shexp.weight q5_K blk.1.ffn_down_exps.weight q5_K +blk.1.ffn_down_shexp.weight q5_K blk.2.attn_qkv.weight q4_K -blk.2.ffn_down_shexp.weight q5_K blk.2.ffn_down_exps.weight q5_K -blk.3.ffn_down_shexp.weight q4_K +blk.2.ffn_down_shexp.weight q5_K blk.3.attn_output.weight q4_K blk.3.attn_v.weight q4_K blk.3.ffn_down_exps.weight q4_K +blk.3.ffn_down_shexp.weight q4_K blk.4.attn_qkv.weight q4_K -blk.4.ffn_down_shexp.weight q4_K blk.4.ffn_down_exps.weight q4_K +blk.4.ffn_down_shexp.weight q4_K blk.5.attn_qkv.weight q4_K -blk.5.ffn_down_shexp.weight q4_K blk.5.ffn_down_exps.weight q4_K +blk.5.ffn_down_shexp.weight q4_K blk.6.attn_qkv.weight q4_K blk.6.ffn_down_exps.weight q4_K blk.6.ffn_down_shexp.weight q4_K -blk.7.ffn_down_shexp.weight q4_K blk.7.attn_output.weight q4_K blk.7.attn_v.weight q4_K blk.7.ffn_down_exps.weight q4_K +blk.7.ffn_down_shexp.weight q4_K blk.8.attn_qkv.weight q4_K -blk.8.ffn_down_shexp.weight q4_K blk.8.ffn_down_exps.weight q4_K +blk.8.ffn_down_shexp.weight q4_K blk.9.attn_qkv.weight q4_K +blk.9.ffn_down_exps.weight q4_K blk.9.ffn_down_shexp.weight q4_K blk.10.attn_qkv.weight q4_K -blk.10.ffn_down_shexp.weight q4_K -blk.9.ffn_down_exps.weight q4_K blk.10.ffn_down_exps.weight q4_K -blk.11.ffn_down_shexp.weight q4_K +blk.10.ffn_down_shexp.weight q4_K blk.11.attn_output.weight q4_K blk.11.attn_v.weight q4_K blk.11.ffn_down_exps.weight q4_K +blk.11.ffn_down_shexp.weight q4_K blk.12.attn_qkv.weight q4_K blk.12.ffn_down_exps.weight q4_K blk.12.ffn_down_shexp.weight q4_K blk.13.attn_qkv.weight q4_K -blk.13.ffn_down_shexp.weight q4_K blk.13.ffn_down_exps.weight q4_K +blk.13.ffn_down_shexp.weight q4_K blk.14.attn_qkv.weight q4_K -blk.14.ffn_down_shexp.weight q4_K blk.14.ffn_down_exps.weight q4_K -blk.15.ffn_down_shexp.weight q4_K +blk.14.ffn_down_shexp.weight q4_K blk.15.attn_output.weight q4_K blk.15.attn_v.weight q4_K blk.15.ffn_down_exps.weight q4_K +blk.15.ffn_down_shexp.weight q4_K blk.16.attn_qkv.weight q4_K -blk.16.ffn_down_shexp.weight q4_K blk.16.ffn_down_exps.weight q4_K +blk.16.ffn_down_shexp.weight q4_K blk.17.attn_qkv.weight q4_K blk.17.ffn_down_exps.weight q4_K blk.17.ffn_down_shexp.weight q4_K blk.18.attn_qkv.weight q4_K -blk.18.ffn_down_shexp.weight q4_K blk.18.ffn_down_exps.weight q4_K -blk.19.ffn_down_shexp.weight q4_K +blk.18.ffn_down_shexp.weight q4_K blk.19.attn_output.weight q4_K blk.19.attn_v.weight q4_K blk.19.ffn_down_exps.weight q4_K +blk.19.ffn_down_shexp.weight q4_K blk.20.attn_qkv.weight q4_K -blk.20.ffn_down_shexp.weight q4_K blk.20.ffn_down_exps.weight q4_K +blk.20.ffn_down_shexp.weight q4_K blk.21.attn_qkv.weight q4_K -blk.21.ffn_down_shexp.weight q4_K blk.21.ffn_down_exps.weight q4_K +blk.21.ffn_down_shexp.weight q4_K blk.22.attn_qkv.weight q4_K -blk.22.ffn_down_shexp.weight q4_K blk.22.ffn_down_exps.weight q4_K -blk.23.ffn_down_exps.weight q4_K -blk.23.ffn_down_shexp.weight q4_K +blk.22.ffn_down_shexp.weight q4_K blk.23.attn_output.weight q4_K blk.23.attn_v.weight q4_K +blk.23.ffn_down_exps.weight q4_K +blk.23.ffn_down_shexp.weight q4_K blk.24.attn_qkv.weight q4_K -blk.24.ffn_down_shexp.weight q4_K blk.24.ffn_down_exps.weight q4_K +blk.24.ffn_down_shexp.weight q4_K blk.25.attn_qkv.weight q4_K -blk.25.ffn_down_shexp.weight q4_K blk.25.ffn_down_exps.weight q4_K +blk.25.ffn_down_shexp.weight q4_K blk.26.attn_qkv.weight q4_K -blk.26.ffn_down_shexp.weight q4_K blk.26.ffn_down_exps.weight q4_K -blk.27.ffn_down_shexp.weight q4_K +blk.26.ffn_down_shexp.weight q4_K blk.27.attn_output.weight q4_K blk.27.attn_v.weight q4_K blk.27.ffn_down_exps.weight q4_K +blk.27.ffn_down_shexp.weight q4_K blk.28.attn_qkv.weight q4_K -blk.28.ffn_down_shexp.weight q4_K blk.28.ffn_down_exps.weight q4_K +blk.28.ffn_down_shexp.weight q4_K blk.29.attn_qkv.weight q4_K blk.29.ffn_down_exps.weight q4_K blk.29.ffn_down_shexp.weight q4_K blk.30.attn_qkv.weight q4_K -blk.30.ffn_down_shexp.weight q4_K blk.30.ffn_down_exps.weight q4_K -blk.31.ffn_down_shexp.weight q4_K +blk.30.ffn_down_shexp.weight q4_K blk.31.attn_output.weight q4_K blk.31.attn_v.weight q4_K blk.31.ffn_down_exps.weight q4_K +blk.31.ffn_down_shexp.weight q4_K blk.32.attn_qkv.weight q4_K -blk.32.ffn_down_shexp.weight q4_K blk.32.ffn_down_exps.weight q4_K +blk.32.ffn_down_shexp.weight q4_K blk.33.attn_qkv.weight q4_K -blk.33.ffn_down_shexp.weight q4_K blk.33.ffn_down_exps.weight q4_K +blk.33.ffn_down_shexp.weight q4_K blk.34.attn_qkv.weight q4_K blk.34.ffn_down_exps.weight q4_K blk.34.ffn_down_shexp.weight q4_K -blk.35.ffn_down_shexp.weight q4_K blk.35.attn_output.weight q4_K blk.35.attn_v.weight q4_K blk.35.ffn_down_exps.weight q4_K +blk.35.ffn_down_shexp.weight q4_K blk.36.attn_qkv.weight q4_K -blk.36.ffn_down_shexp.weight q4_K blk.36.ffn_down_exps.weight q4_K +blk.36.ffn_down_shexp.weight q4_K blk.37.attn_qkv.weight q4_K -blk.37.ffn_down_shexp.weight q4_K blk.37.ffn_down_exps.weight q4_K +blk.37.ffn_down_shexp.weight q4_K blk.38.attn_qkv.weight q4_K -blk.38.ffn_down_shexp.weight q4_K blk.38.ffn_down_exps.weight q4_K -blk.39.ffn_down_shexp.weight q4_K +blk.38.ffn_down_shexp.weight q4_K blk.39.attn_output.weight q4_K blk.39.attn_v.weight q4_K blk.39.ffn_down_exps.weight q4_K +blk.39.ffn_down_shexp.weight q4_K blk.40.attn_qkv.weight q4_K blk.40.ffn_down_exps.weight q4_K blk.40.ffn_down_shexp.weight q4_K blk.41.attn_qkv.weight q4_K -blk.41.ffn_down_shexp.weight q4_K blk.41.ffn_down_exps.weight q4_K +blk.41.ffn_down_shexp.weight q4_K blk.42.attn_qkv.weight q4_K -blk.42.ffn_down_shexp.weight q4_K blk.42.ffn_down_exps.weight q4_K -blk.43.ffn_down_shexp.weight q4_K +blk.42.ffn_down_shexp.weight q4_K blk.43.attn_output.weight q4_K blk.43.attn_v.weight q4_K blk.43.ffn_down_exps.weight q4_K +blk.43.ffn_down_shexp.weight q4_K blk.44.attn_qkv.weight q4_K -blk.44.ffn_down_shexp.weight q4_K blk.44.ffn_down_exps.weight q4_K +blk.44.ffn_down_shexp.weight q4_K blk.45.attn_qkv.weight q4_K -blk.45.ffn_down_shexp.weight q4_K blk.45.ffn_down_exps.weight q4_K +blk.45.ffn_down_shexp.weight q4_K blk.46.attn_qkv.weight q4_K blk.46.ffn_down_exps.weight q4_K blk.46.ffn_down_shexp.weight q4_K -blk.47.ffn_down_shexp.weight q4_K blk.47.attn_output.weight q4_K blk.47.attn_v.weight q4_K -output.weight q6_K blk.47.ffn_down_exps.weight q4_K +blk.47.ffn_down_shexp.weight q4_K [Q3_K_L] q3_K +output.weight q6_K blk.0.attn_qkv.weight q5_K blk.0.ffn_down_exps.weight q5_K blk.0.ffn_down_shexp.weight q5_K blk.1.attn_qkv.weight q5_K -blk.1.ffn_down_shexp.weight q5_K blk.1.ffn_down_exps.weight q5_K +blk.1.ffn_down_shexp.weight q5_K blk.2.attn_qkv.weight q5_K -blk.2.ffn_down_shexp.weight q5_K blk.2.ffn_down_exps.weight q5_K -blk.3.ffn_down_shexp.weight q5_K +blk.2.ffn_down_shexp.weight q5_K blk.3.attn_output.weight q5_K blk.3.attn_v.weight q5_K blk.3.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q5_K blk.4.attn_qkv.weight q5_K -blk.4.ffn_down_shexp.weight q5_K blk.4.ffn_down_exps.weight q5_K +blk.4.ffn_down_shexp.weight q5_K blk.5.attn_qkv.weight q5_K -blk.5.ffn_down_shexp.weight q5_K blk.5.ffn_down_exps.weight q5_K +blk.5.ffn_down_shexp.weight q5_K blk.6.attn_qkv.weight q5_K blk.6.ffn_down_exps.weight q5_K blk.6.ffn_down_shexp.weight q5_K -blk.7.ffn_down_shexp.weight q5_K blk.7.attn_output.weight q5_K blk.7.attn_v.weight q5_K blk.7.ffn_down_exps.weight q5_K +blk.7.ffn_down_shexp.weight q5_K blk.8.attn_qkv.weight q5_K -blk.8.ffn_down_shexp.weight q5_K blk.8.ffn_down_exps.weight q5_K +blk.8.ffn_down_shexp.weight q5_K blk.9.attn_qkv.weight q5_K +blk.9.ffn_down_exps.weight q5_K blk.9.ffn_down_shexp.weight q5_K blk.10.attn_qkv.weight q5_K -blk.10.ffn_down_shexp.weight q5_K -blk.9.ffn_down_exps.weight q5_K blk.10.ffn_down_exps.weight q5_K -blk.11.ffn_down_shexp.weight q5_K +blk.10.ffn_down_shexp.weight q5_K blk.11.attn_output.weight q5_K blk.11.attn_v.weight q5_K blk.11.ffn_down_exps.weight q5_K +blk.11.ffn_down_shexp.weight q5_K blk.12.attn_qkv.weight q5_K blk.12.ffn_down_exps.weight q5_K blk.12.ffn_down_shexp.weight q5_K blk.13.attn_qkv.weight q5_K -blk.13.ffn_down_shexp.weight q5_K blk.13.ffn_down_exps.weight q5_K +blk.13.ffn_down_shexp.weight q5_K blk.14.attn_qkv.weight q5_K -blk.14.ffn_down_shexp.weight q5_K blk.14.ffn_down_exps.weight q5_K -blk.15.ffn_down_shexp.weight q5_K +blk.14.ffn_down_shexp.weight q5_K blk.15.attn_output.weight q5_K blk.15.attn_v.weight q5_K blk.15.ffn_down_exps.weight q5_K +blk.15.ffn_down_shexp.weight q5_K blk.16.attn_qkv.weight q5_K -blk.16.ffn_down_shexp.weight q5_K blk.16.ffn_down_exps.weight q5_K +blk.16.ffn_down_shexp.weight q5_K blk.17.attn_qkv.weight q5_K blk.17.ffn_down_exps.weight q5_K blk.17.ffn_down_shexp.weight q5_K blk.18.attn_qkv.weight q5_K -blk.18.ffn_down_shexp.weight q5_K blk.18.ffn_down_exps.weight q5_K -blk.19.ffn_down_shexp.weight q5_K +blk.18.ffn_down_shexp.weight q5_K blk.19.attn_output.weight q5_K blk.19.attn_v.weight q5_K blk.19.ffn_down_exps.weight q5_K +blk.19.ffn_down_shexp.weight q5_K blk.20.attn_qkv.weight q5_K -blk.20.ffn_down_shexp.weight q5_K blk.20.ffn_down_exps.weight q5_K +blk.20.ffn_down_shexp.weight q5_K blk.21.attn_qkv.weight q5_K -blk.21.ffn_down_shexp.weight q5_K blk.21.ffn_down_exps.weight q5_K +blk.21.ffn_down_shexp.weight q5_K blk.22.attn_qkv.weight q5_K -blk.22.ffn_down_shexp.weight q5_K blk.22.ffn_down_exps.weight q5_K -blk.23.ffn_down_exps.weight q5_K -blk.23.ffn_down_shexp.weight q5_K +blk.22.ffn_down_shexp.weight q5_K blk.23.attn_output.weight q5_K blk.23.attn_v.weight q5_K +blk.23.ffn_down_exps.weight q5_K +blk.23.ffn_down_shexp.weight q5_K blk.24.attn_qkv.weight q5_K -blk.24.ffn_down_shexp.weight q5_K blk.24.ffn_down_exps.weight q5_K +blk.24.ffn_down_shexp.weight q5_K blk.25.attn_qkv.weight q5_K -blk.25.ffn_down_shexp.weight q5_K blk.25.ffn_down_exps.weight q5_K +blk.25.ffn_down_shexp.weight q5_K blk.26.attn_qkv.weight q5_K -blk.26.ffn_down_shexp.weight q5_K blk.26.ffn_down_exps.weight q5_K -blk.27.ffn_down_shexp.weight q5_K +blk.26.ffn_down_shexp.weight q5_K blk.27.attn_output.weight q5_K blk.27.attn_v.weight q5_K blk.27.ffn_down_exps.weight q5_K +blk.27.ffn_down_shexp.weight q5_K blk.28.attn_qkv.weight q5_K -blk.28.ffn_down_shexp.weight q5_K blk.28.ffn_down_exps.weight q5_K +blk.28.ffn_down_shexp.weight q5_K blk.29.attn_qkv.weight q5_K blk.29.ffn_down_exps.weight q5_K blk.29.ffn_down_shexp.weight q5_K blk.30.attn_qkv.weight q5_K -blk.30.ffn_down_shexp.weight q5_K blk.30.ffn_down_exps.weight q5_K -blk.31.ffn_down_shexp.weight q5_K +blk.30.ffn_down_shexp.weight q5_K blk.31.attn_output.weight q5_K blk.31.attn_v.weight q5_K blk.31.ffn_down_exps.weight q5_K +blk.31.ffn_down_shexp.weight q5_K blk.32.attn_qkv.weight q5_K -blk.32.ffn_down_shexp.weight q5_K blk.32.ffn_down_exps.weight q5_K +blk.32.ffn_down_shexp.weight q5_K blk.33.attn_qkv.weight q5_K -blk.33.ffn_down_shexp.weight q5_K blk.33.ffn_down_exps.weight q5_K +blk.33.ffn_down_shexp.weight q5_K blk.34.attn_qkv.weight q5_K blk.34.ffn_down_exps.weight q5_K blk.34.ffn_down_shexp.weight q5_K -blk.35.ffn_down_shexp.weight q5_K blk.35.attn_output.weight q5_K blk.35.attn_v.weight q5_K blk.35.ffn_down_exps.weight q5_K +blk.35.ffn_down_shexp.weight q5_K blk.36.attn_qkv.weight q5_K -blk.36.ffn_down_shexp.weight q5_K blk.36.ffn_down_exps.weight q5_K +blk.36.ffn_down_shexp.weight q5_K blk.37.attn_qkv.weight q5_K -blk.37.ffn_down_shexp.weight q5_K blk.37.ffn_down_exps.weight q5_K +blk.37.ffn_down_shexp.weight q5_K blk.38.attn_qkv.weight q5_K -blk.38.ffn_down_shexp.weight q5_K blk.38.ffn_down_exps.weight q5_K -blk.39.ffn_down_shexp.weight q5_K +blk.38.ffn_down_shexp.weight q5_K blk.39.attn_output.weight q5_K blk.39.attn_v.weight q5_K blk.39.ffn_down_exps.weight q5_K +blk.39.ffn_down_shexp.weight q5_K blk.40.attn_qkv.weight q5_K blk.40.ffn_down_exps.weight q5_K blk.40.ffn_down_shexp.weight q5_K blk.41.attn_qkv.weight q5_K -blk.41.ffn_down_shexp.weight q5_K blk.41.ffn_down_exps.weight q5_K +blk.41.ffn_down_shexp.weight q5_K blk.42.attn_qkv.weight q5_K -blk.42.ffn_down_shexp.weight q5_K blk.42.ffn_down_exps.weight q5_K -blk.43.ffn_down_shexp.weight q5_K +blk.42.ffn_down_shexp.weight q5_K blk.43.attn_output.weight q5_K blk.43.attn_v.weight q5_K blk.43.ffn_down_exps.weight q5_K +blk.43.ffn_down_shexp.weight q5_K blk.44.attn_qkv.weight q5_K -blk.44.ffn_down_shexp.weight q5_K blk.44.ffn_down_exps.weight q5_K +blk.44.ffn_down_shexp.weight q5_K blk.45.attn_qkv.weight q5_K -blk.45.ffn_down_shexp.weight q5_K blk.45.ffn_down_exps.weight q5_K +blk.45.ffn_down_shexp.weight q5_K blk.46.attn_qkv.weight q5_K blk.46.ffn_down_exps.weight q5_K blk.46.ffn_down_shexp.weight q5_K -blk.47.ffn_down_shexp.weight q5_K blk.47.attn_output.weight q5_K blk.47.attn_v.weight q5_K -output.weight q6_K blk.47.ffn_down_exps.weight q5_K +blk.47.ffn_down_shexp.weight q5_K [Q4_K_S] q4_K +output.weight q6_K blk.0.attn_qkv.weight q5_K blk.0.ffn_down_exps.weight q5_K blk.0.ffn_down_shexp.weight q5_K blk.1.attn_qkv.weight q5_K -blk.1.ffn_down_shexp.weight q5_K blk.1.ffn_down_exps.weight q5_K +blk.1.ffn_down_shexp.weight q5_K blk.2.attn_qkv.weight q5_K -blk.2.ffn_down_shexp.weight q5_K blk.2.ffn_down_exps.weight q5_K -blk.3.ffn_down_shexp.weight q5_K +blk.2.ffn_down_shexp.weight q5_K blk.3.attn_v.weight q5_K blk.3.ffn_down_exps.weight q5_K -blk.4.ffn_down_shexp.weight q5_K +blk.3.ffn_down_shexp.weight q5_K blk.4.ffn_down_exps.weight q5_K -blk.5.ffn_down_shexp.weight q5_K +blk.4.ffn_down_shexp.weight q5_K blk.5.ffn_down_exps.weight q5_K -output.weight q6_K +blk.5.ffn_down_shexp.weight q5_K [Q4_K_M] q4_K +output.weight q6_K blk.0.attn_qkv.weight q6_K blk.0.ffn_down_exps.weight q6_K blk.0.ffn_down_shexp.weight q6_K blk.1.attn_qkv.weight q6_K -blk.1.ffn_down_shexp.weight q6_K blk.1.ffn_down_exps.weight q6_K +blk.1.ffn_down_shexp.weight q6_K blk.2.attn_qkv.weight q6_K -blk.2.ffn_down_shexp.weight q6_K blk.2.ffn_down_exps.weight q6_K -blk.3.ffn_down_shexp.weight q6_K +blk.2.ffn_down_shexp.weight q6_K blk.3.attn_v.weight q6_K blk.3.ffn_down_exps.weight q6_K +blk.3.ffn_down_shexp.weight q6_K blk.4.attn_qkv.weight q6_K -blk.4.ffn_down_shexp.weight q6_K blk.4.ffn_down_exps.weight q6_K +blk.4.ffn_down_shexp.weight q6_K blk.5.attn_qkv.weight q6_K -blk.5.ffn_down_shexp.weight q6_K blk.5.ffn_down_exps.weight q6_K +blk.5.ffn_down_shexp.weight q6_K blk.8.attn_qkv.weight q6_K -blk.8.ffn_down_shexp.weight q6_K blk.8.ffn_down_exps.weight q6_K -blk.11.ffn_down_shexp.weight q6_K +blk.8.ffn_down_shexp.weight q6_K blk.11.attn_v.weight q6_K blk.11.ffn_down_exps.weight q6_K +blk.11.ffn_down_shexp.weight q6_K blk.14.attn_qkv.weight q6_K -blk.14.ffn_down_shexp.weight q6_K blk.14.ffn_down_exps.weight q6_K +blk.14.ffn_down_shexp.weight q6_K blk.17.attn_qkv.weight q6_K blk.17.ffn_down_exps.weight q6_K blk.17.ffn_down_shexp.weight q6_K blk.20.attn_qkv.weight q6_K -blk.20.ffn_down_shexp.weight q6_K blk.20.ffn_down_exps.weight q6_K +blk.20.ffn_down_shexp.weight q6_K +blk.23.attn_v.weight q6_K blk.23.ffn_down_exps.weight q6_K blk.23.ffn_down_shexp.weight q6_K -blk.23.attn_v.weight q6_K blk.26.attn_qkv.weight q6_K -blk.26.ffn_down_shexp.weight q6_K blk.26.ffn_down_exps.weight q6_K +blk.26.ffn_down_shexp.weight q6_K blk.29.attn_qkv.weight q6_K blk.29.ffn_down_exps.weight q6_K blk.29.ffn_down_shexp.weight q6_K blk.32.attn_qkv.weight q6_K -blk.32.ffn_down_shexp.weight q6_K blk.32.ffn_down_exps.weight q6_K -blk.35.ffn_down_shexp.weight q6_K +blk.32.ffn_down_shexp.weight q6_K blk.35.attn_v.weight q6_K blk.35.ffn_down_exps.weight q6_K +blk.35.ffn_down_shexp.weight q6_K blk.38.attn_qkv.weight q6_K -blk.38.ffn_down_shexp.weight q6_K blk.38.ffn_down_exps.weight q6_K +blk.38.ffn_down_shexp.weight q6_K blk.41.attn_qkv.weight q6_K -blk.41.ffn_down_shexp.weight q6_K blk.41.ffn_down_exps.weight q6_K +blk.41.ffn_down_shexp.weight q6_K blk.42.attn_qkv.weight q6_K -blk.42.ffn_down_shexp.weight q6_K blk.42.ffn_down_exps.weight q6_K -blk.43.ffn_down_shexp.weight q6_K +blk.42.ffn_down_shexp.weight q6_K blk.43.attn_v.weight q6_K blk.43.ffn_down_exps.weight q6_K +blk.43.ffn_down_shexp.weight q6_K blk.44.attn_qkv.weight q6_K -blk.44.ffn_down_shexp.weight q6_K blk.44.ffn_down_exps.weight q6_K +blk.44.ffn_down_shexp.weight q6_K blk.45.attn_qkv.weight q6_K -blk.45.ffn_down_shexp.weight q6_K blk.45.ffn_down_exps.weight q6_K +blk.45.ffn_down_shexp.weight q6_K blk.46.attn_qkv.weight q6_K blk.46.ffn_down_exps.weight q6_K blk.46.ffn_down_shexp.weight q6_K -blk.47.ffn_down_shexp.weight q6_K blk.47.attn_v.weight q6_K -output.weight q6_K blk.47.ffn_down_exps.weight q6_K +blk.47.ffn_down_shexp.weight q6_K [Q5_K_S] q5_K output.weight q6_K [Q5_K_M] q5_K +output.weight q6_K blk.0.attn_qkv.weight q6_K blk.0.ffn_down_exps.weight q6_K blk.0.ffn_down_shexp.weight q6_K blk.1.attn_qkv.weight q6_K -blk.1.ffn_down_shexp.weight q6_K blk.1.ffn_down_exps.weight q6_K +blk.1.ffn_down_shexp.weight q6_K blk.2.attn_qkv.weight q6_K -blk.2.ffn_down_shexp.weight q6_K blk.2.ffn_down_exps.weight q6_K -blk.3.ffn_down_shexp.weight q6_K +blk.2.ffn_down_shexp.weight q6_K blk.3.attn_v.weight q6_K blk.3.ffn_down_exps.weight q6_K +blk.3.ffn_down_shexp.weight q6_K blk.4.attn_qkv.weight q6_K -blk.4.ffn_down_shexp.weight q6_K blk.4.ffn_down_exps.weight q6_K +blk.4.ffn_down_shexp.weight q6_K blk.5.attn_qkv.weight q6_K -blk.5.ffn_down_shexp.weight q6_K blk.5.ffn_down_exps.weight q6_K +blk.5.ffn_down_shexp.weight q6_K blk.8.attn_qkv.weight q6_K -blk.8.ffn_down_shexp.weight q6_K blk.8.ffn_down_exps.weight q6_K -blk.11.ffn_down_shexp.weight q6_K +blk.8.ffn_down_shexp.weight q6_K blk.11.attn_v.weight q6_K blk.11.ffn_down_exps.weight q6_K +blk.11.ffn_down_shexp.weight q6_K blk.14.attn_qkv.weight q6_K -blk.14.ffn_down_shexp.weight q6_K blk.14.ffn_down_exps.weight q6_K +blk.14.ffn_down_shexp.weight q6_K blk.17.attn_qkv.weight q6_K blk.17.ffn_down_exps.weight q6_K blk.17.ffn_down_shexp.weight q6_K blk.20.attn_qkv.weight q6_K -blk.20.ffn_down_shexp.weight q6_K blk.20.ffn_down_exps.weight q6_K +blk.20.ffn_down_shexp.weight q6_K +blk.23.attn_v.weight q6_K blk.23.ffn_down_exps.weight q6_K blk.23.ffn_down_shexp.weight q6_K -blk.23.attn_v.weight q6_K blk.26.attn_qkv.weight q6_K -blk.26.ffn_down_shexp.weight q6_K blk.26.ffn_down_exps.weight q6_K +blk.26.ffn_down_shexp.weight q6_K blk.29.attn_qkv.weight q6_K blk.29.ffn_down_exps.weight q6_K blk.29.ffn_down_shexp.weight q6_K blk.32.attn_qkv.weight q6_K -blk.32.ffn_down_shexp.weight q6_K blk.32.ffn_down_exps.weight q6_K -blk.35.ffn_down_shexp.weight q6_K +blk.32.ffn_down_shexp.weight q6_K blk.35.attn_v.weight q6_K blk.35.ffn_down_exps.weight q6_K +blk.35.ffn_down_shexp.weight q6_K blk.38.attn_qkv.weight q6_K -blk.38.ffn_down_shexp.weight q6_K blk.38.ffn_down_exps.weight q6_K +blk.38.ffn_down_shexp.weight q6_K blk.41.attn_qkv.weight q6_K -blk.41.ffn_down_shexp.weight q6_K blk.41.ffn_down_exps.weight q6_K +blk.41.ffn_down_shexp.weight q6_K blk.42.attn_qkv.weight q6_K -blk.42.ffn_down_shexp.weight q6_K blk.42.ffn_down_exps.weight q6_K -blk.43.ffn_down_shexp.weight q6_K +blk.42.ffn_down_shexp.weight q6_K blk.43.attn_v.weight q6_K blk.43.ffn_down_exps.weight q6_K +blk.43.ffn_down_shexp.weight q6_K blk.44.attn_qkv.weight q6_K -blk.44.ffn_down_shexp.weight q6_K blk.44.ffn_down_exps.weight q6_K +blk.44.ffn_down_shexp.weight q6_K blk.45.attn_qkv.weight q6_K -blk.45.ffn_down_shexp.weight q6_K blk.45.ffn_down_exps.weight q6_K +blk.45.ffn_down_shexp.weight q6_K blk.46.attn_qkv.weight q6_K blk.46.ffn_down_exps.weight q6_K blk.46.ffn_down_shexp.weight q6_K -blk.47.ffn_down_shexp.weight q6_K blk.47.attn_v.weight q6_K -output.weight q6_K blk.47.ffn_down_exps.weight q6_K +blk.47.ffn_down_shexp.weight q6_K [Q6_K] q6_K [IQ2_XXS] iq2_xxs +output.weight q5_K token_embd.weight q2_K blk.0.attn_qkv.weight q4_K blk.0.ffn_down_exps.weight q2_K blk.0.ffn_down_shexp.weight q2_K blk.1.attn_qkv.weight q4_K -blk.1.ffn_down_shexp.weight q2_K blk.1.ffn_down_exps.weight q2_K +blk.1.ffn_down_shexp.weight q2_K blk.2.attn_qkv.weight q4_K -blk.2.ffn_down_shexp.weight q2_K blk.2.ffn_down_exps.weight q2_K +blk.2.ffn_down_shexp.weight q2_K blk.3.attn_v.weight q4_K blk.4.attn_qkv.weight q4_K blk.5.attn_qkv.weight q4_K @@ -729,19 +730,19 @@ blk.44.attn_qkv.weight q4_K blk.45.attn_qkv.weight q4_K blk.46.attn_qkv.weight q4_K blk.47.attn_v.weight q4_K -output.weight q5_K [IQ2_XS] iq2_xs +output.weight q5_K token_embd.weight q2_K blk.0.attn_qkv.weight q4_K blk.0.ffn_down_exps.weight q2_K blk.0.ffn_down_shexp.weight q2_K blk.1.attn_qkv.weight q4_K -blk.1.ffn_down_shexp.weight q2_K blk.1.ffn_down_exps.weight q2_K +blk.1.ffn_down_shexp.weight q2_K blk.2.attn_qkv.weight q4_K -blk.2.ffn_down_shexp.weight q2_K blk.2.ffn_down_exps.weight q2_K +blk.2.ffn_down_shexp.weight q2_K blk.3.attn_v.weight q4_K blk.4.attn_qkv.weight q4_K blk.5.attn_qkv.weight q4_K @@ -787,27 +788,27 @@ blk.44.attn_qkv.weight q4_K blk.45.attn_qkv.weight q4_K blk.46.attn_qkv.weight q4_K blk.47.attn_v.weight q4_K -output.weight q5_K [Q2_K_S] q2_K +output.weight q6_K blk.0.attn_qkv.weight q4_K blk.0.ffn_down_exps.weight q4_K blk.0.ffn_down_shexp.weight q4_K blk.1.attn_qkv.weight q4_K -blk.1.ffn_down_shexp.weight q4_K blk.1.ffn_down_exps.weight q4_K +blk.1.ffn_down_shexp.weight q4_K blk.2.attn_qkv.weight q4_K -blk.2.ffn_down_shexp.weight q4_K blk.2.ffn_down_exps.weight q4_K -blk.3.ffn_down_shexp.weight q4_K +blk.2.ffn_down_shexp.weight q4_K blk.3.attn_v.weight q4_K blk.3.ffn_down_exps.weight q4_K +blk.3.ffn_down_shexp.weight q4_K blk.4.attn_qkv.weight q4_K -blk.4.ffn_down_shexp.weight q4_K blk.4.ffn_down_exps.weight q4_K +blk.4.ffn_down_shexp.weight q4_K blk.5.attn_qkv.weight q4_K -blk.5.ffn_down_shexp.weight q4_K blk.5.ffn_down_exps.weight q4_K +blk.5.ffn_down_shexp.weight q4_K blk.6.attn_qkv.weight q4_K blk.7.attn_v.weight q4_K blk.8.attn_qkv.weight q4_K @@ -850,9 +851,9 @@ blk.44.attn_qkv.weight q4_K blk.45.attn_qkv.weight q4_K blk.46.attn_qkv.weight q4_K blk.47.attn_v.weight q4_K -output.weight q6_K [IQ3_XS] iq3_s +output.weight q6_K blk.0.attn_qkv.weight q4_K blk.1.attn_qkv.weight q4_K blk.2.attn_qkv.weight q4_K @@ -863,202 +864,202 @@ blk.4.attn_qkv.weight q4_K blk.5.attn_qkv.weight q4_K blk.6.attn_qkv.weight q4_K blk.6.ffn_gate_exps.weight iq3_xxs -blk.6.ffn_up_exps.weight iq3_xxs blk.6.ffn_gate_shexp.weight iq3_xxs +blk.6.ffn_up_exps.weight iq3_xxs blk.6.ffn_up_shexp.weight iq3_xxs -blk.7.ffn_gate_shexp.weight iq3_xxs -blk.7.ffn_up_shexp.weight iq3_xxs blk.7.attn_k.weight iq3_xxs blk.7.attn_q.weight iq3_xxs blk.7.attn_v.weight q4_K blk.7.ffn_gate_exps.weight iq3_xxs +blk.7.ffn_gate_shexp.weight iq3_xxs blk.7.ffn_up_exps.weight iq3_xxs +blk.7.ffn_up_shexp.weight iq3_xxs blk.8.attn_qkv.weight q4_K -blk.8.ffn_gate_shexp.weight iq3_xxs -blk.8.ffn_up_shexp.weight iq3_xxs blk.8.ffn_gate_exps.weight iq3_xxs +blk.8.ffn_gate_shexp.weight iq3_xxs blk.8.ffn_up_exps.weight iq3_xxs +blk.8.ffn_up_shexp.weight iq3_xxs blk.9.attn_qkv.weight q4_K +blk.9.ffn_gate_exps.weight iq3_xxs blk.9.ffn_gate_shexp.weight iq3_xxs +blk.9.ffn_up_exps.weight iq3_xxs blk.9.ffn_up_shexp.weight iq3_xxs blk.10.attn_qkv.weight q4_K -blk.10.ffn_gate_shexp.weight iq3_xxs -blk.10.ffn_up_shexp.weight iq3_xxs -blk.9.ffn_gate_exps.weight iq3_xxs -blk.9.ffn_up_exps.weight iq3_xxs blk.10.ffn_gate_exps.weight iq3_xxs +blk.10.ffn_gate_shexp.weight iq3_xxs blk.10.ffn_up_exps.weight iq3_xxs -blk.11.ffn_gate_shexp.weight iq3_xxs -blk.11.ffn_up_shexp.weight iq3_xxs +blk.10.ffn_up_shexp.weight iq3_xxs blk.11.attn_k.weight iq3_xxs blk.11.attn_q.weight iq3_xxs blk.11.attn_v.weight q4_K blk.11.ffn_gate_exps.weight iq3_xxs +blk.11.ffn_gate_shexp.weight iq3_xxs blk.11.ffn_up_exps.weight iq3_xxs +blk.11.ffn_up_shexp.weight iq3_xxs blk.12.attn_qkv.weight q4_K blk.12.ffn_gate_exps.weight iq3_xxs -blk.12.ffn_up_exps.weight iq3_xxs blk.12.ffn_gate_shexp.weight iq3_xxs +blk.12.ffn_up_exps.weight iq3_xxs blk.12.ffn_up_shexp.weight iq3_xxs blk.13.attn_qkv.weight q4_K -blk.13.ffn_gate_shexp.weight iq3_xxs -blk.13.ffn_up_shexp.weight iq3_xxs blk.13.ffn_gate_exps.weight iq3_xxs +blk.13.ffn_gate_shexp.weight iq3_xxs blk.13.ffn_up_exps.weight iq3_xxs +blk.13.ffn_up_shexp.weight iq3_xxs blk.14.attn_qkv.weight q4_K -blk.14.ffn_gate_shexp.weight iq3_xxs -blk.14.ffn_up_shexp.weight iq3_xxs blk.14.ffn_gate_exps.weight iq3_xxs +blk.14.ffn_gate_shexp.weight iq3_xxs blk.14.ffn_up_exps.weight iq3_xxs -blk.15.ffn_gate_shexp.weight iq3_xxs -blk.15.ffn_up_shexp.weight iq3_xxs +blk.14.ffn_up_shexp.weight iq3_xxs blk.15.attn_k.weight iq3_xxs blk.15.attn_q.weight iq3_xxs blk.15.attn_v.weight q4_K blk.15.ffn_gate_exps.weight iq3_xxs +blk.15.ffn_gate_shexp.weight iq3_xxs blk.15.ffn_up_exps.weight iq3_xxs +blk.15.ffn_up_shexp.weight iq3_xxs blk.16.attn_qkv.weight q4_K -blk.16.ffn_gate_shexp.weight iq3_xxs -blk.16.ffn_up_shexp.weight iq3_xxs blk.16.ffn_gate_exps.weight iq3_xxs +blk.16.ffn_gate_shexp.weight iq3_xxs blk.16.ffn_up_exps.weight iq3_xxs +blk.16.ffn_up_shexp.weight iq3_xxs blk.17.attn_qkv.weight q4_K blk.17.ffn_gate_exps.weight iq3_xxs -blk.17.ffn_up_exps.weight iq3_xxs blk.17.ffn_gate_shexp.weight iq3_xxs +blk.17.ffn_up_exps.weight iq3_xxs blk.17.ffn_up_shexp.weight iq3_xxs blk.18.attn_qkv.weight q4_K -blk.18.ffn_gate_shexp.weight iq3_xxs -blk.18.ffn_up_shexp.weight iq3_xxs blk.18.ffn_gate_exps.weight iq3_xxs +blk.18.ffn_gate_shexp.weight iq3_xxs blk.18.ffn_up_exps.weight iq3_xxs -blk.19.ffn_gate_shexp.weight iq3_xxs -blk.19.ffn_up_shexp.weight iq3_xxs +blk.18.ffn_up_shexp.weight iq3_xxs blk.19.attn_k.weight iq3_xxs blk.19.attn_q.weight iq3_xxs blk.19.attn_v.weight q4_K blk.19.ffn_gate_exps.weight iq3_xxs +blk.19.ffn_gate_shexp.weight iq3_xxs blk.19.ffn_up_exps.weight iq3_xxs +blk.19.ffn_up_shexp.weight iq3_xxs blk.20.attn_qkv.weight q4_K -blk.20.ffn_gate_shexp.weight iq3_xxs -blk.20.ffn_up_shexp.weight iq3_xxs blk.20.ffn_gate_exps.weight iq3_xxs +blk.20.ffn_gate_shexp.weight iq3_xxs blk.20.ffn_up_exps.weight iq3_xxs +blk.20.ffn_up_shexp.weight iq3_xxs blk.21.attn_qkv.weight q4_K -blk.21.ffn_gate_shexp.weight iq3_xxs -blk.21.ffn_up_shexp.weight iq3_xxs blk.21.ffn_gate_exps.weight iq3_xxs +blk.21.ffn_gate_shexp.weight iq3_xxs blk.21.ffn_up_exps.weight iq3_xxs +blk.21.ffn_up_shexp.weight iq3_xxs blk.22.attn_qkv.weight q4_K -blk.22.ffn_gate_shexp.weight iq3_xxs -blk.22.ffn_up_shexp.weight iq3_xxs blk.22.ffn_gate_exps.weight iq3_xxs +blk.22.ffn_gate_shexp.weight iq3_xxs blk.22.ffn_up_exps.weight iq3_xxs -blk.23.ffn_gate_exps.weight iq3_xxs -blk.23.ffn_up_exps.weight iq3_xxs -blk.23.ffn_gate_shexp.weight iq3_xxs -blk.23.ffn_up_shexp.weight iq3_xxs +blk.22.ffn_up_shexp.weight iq3_xxs blk.23.attn_k.weight iq3_xxs blk.23.attn_q.weight iq3_xxs blk.23.attn_v.weight q4_K +blk.23.ffn_gate_exps.weight iq3_xxs +blk.23.ffn_gate_shexp.weight iq3_xxs +blk.23.ffn_up_exps.weight iq3_xxs +blk.23.ffn_up_shexp.weight iq3_xxs blk.24.attn_qkv.weight q4_K -blk.24.ffn_gate_shexp.weight iq3_xxs -blk.24.ffn_up_shexp.weight iq3_xxs blk.24.ffn_gate_exps.weight iq3_xxs +blk.24.ffn_gate_shexp.weight iq3_xxs blk.24.ffn_up_exps.weight iq3_xxs +blk.24.ffn_up_shexp.weight iq3_xxs blk.25.attn_qkv.weight q4_K -blk.25.ffn_gate_shexp.weight iq3_xxs -blk.25.ffn_up_shexp.weight iq3_xxs blk.25.ffn_gate_exps.weight iq3_xxs +blk.25.ffn_gate_shexp.weight iq3_xxs blk.25.ffn_up_exps.weight iq3_xxs +blk.25.ffn_up_shexp.weight iq3_xxs blk.26.attn_qkv.weight q4_K -blk.26.ffn_gate_shexp.weight iq3_xxs -blk.26.ffn_up_shexp.weight iq3_xxs blk.26.ffn_gate_exps.weight iq3_xxs +blk.26.ffn_gate_shexp.weight iq3_xxs blk.26.ffn_up_exps.weight iq3_xxs -blk.27.ffn_gate_shexp.weight iq3_xxs -blk.27.ffn_up_shexp.weight iq3_xxs +blk.26.ffn_up_shexp.weight iq3_xxs blk.27.attn_k.weight iq3_xxs blk.27.attn_q.weight iq3_xxs blk.27.attn_v.weight q4_K blk.27.ffn_gate_exps.weight iq3_xxs +blk.27.ffn_gate_shexp.weight iq3_xxs blk.27.ffn_up_exps.weight iq3_xxs +blk.27.ffn_up_shexp.weight iq3_xxs blk.28.attn_qkv.weight q4_K -blk.28.ffn_gate_shexp.weight iq3_xxs -blk.28.ffn_up_shexp.weight iq3_xxs blk.28.ffn_gate_exps.weight iq3_xxs +blk.28.ffn_gate_shexp.weight iq3_xxs blk.28.ffn_up_exps.weight iq3_xxs +blk.28.ffn_up_shexp.weight iq3_xxs blk.29.attn_qkv.weight q4_K blk.29.ffn_gate_exps.weight iq3_xxs -blk.29.ffn_up_exps.weight iq3_xxs blk.29.ffn_gate_shexp.weight iq3_xxs +blk.29.ffn_up_exps.weight iq3_xxs blk.29.ffn_up_shexp.weight iq3_xxs blk.30.attn_qkv.weight q4_K -blk.30.ffn_gate_shexp.weight iq3_xxs -blk.30.ffn_up_shexp.weight iq3_xxs blk.30.ffn_gate_exps.weight iq3_xxs +blk.30.ffn_gate_shexp.weight iq3_xxs blk.30.ffn_up_exps.weight iq3_xxs -blk.31.ffn_gate_shexp.weight iq3_xxs -blk.31.ffn_up_shexp.weight iq3_xxs +blk.30.ffn_up_shexp.weight iq3_xxs blk.31.attn_k.weight iq3_xxs blk.31.attn_q.weight iq3_xxs blk.31.attn_v.weight q4_K blk.31.ffn_gate_exps.weight iq3_xxs +blk.31.ffn_gate_shexp.weight iq3_xxs blk.31.ffn_up_exps.weight iq3_xxs +blk.31.ffn_up_shexp.weight iq3_xxs blk.32.attn_qkv.weight q4_K -blk.32.ffn_gate_shexp.weight iq3_xxs -blk.32.ffn_up_shexp.weight iq3_xxs blk.32.ffn_gate_exps.weight iq3_xxs +blk.32.ffn_gate_shexp.weight iq3_xxs blk.32.ffn_up_exps.weight iq3_xxs +blk.32.ffn_up_shexp.weight iq3_xxs blk.33.attn_qkv.weight q4_K -blk.33.ffn_gate_shexp.weight iq3_xxs -blk.33.ffn_up_shexp.weight iq3_xxs blk.33.ffn_gate_exps.weight iq3_xxs +blk.33.ffn_gate_shexp.weight iq3_xxs blk.33.ffn_up_exps.weight iq3_xxs +blk.33.ffn_up_shexp.weight iq3_xxs blk.34.attn_qkv.weight q4_K blk.34.ffn_gate_exps.weight iq3_xxs -blk.34.ffn_up_exps.weight iq3_xxs blk.34.ffn_gate_shexp.weight iq3_xxs +blk.34.ffn_up_exps.weight iq3_xxs blk.34.ffn_up_shexp.weight iq3_xxs -blk.35.ffn_gate_shexp.weight iq3_xxs -blk.35.ffn_up_shexp.weight iq3_xxs blk.35.attn_k.weight iq3_xxs blk.35.attn_q.weight iq3_xxs blk.35.attn_v.weight q4_K blk.35.ffn_gate_exps.weight iq3_xxs +blk.35.ffn_gate_shexp.weight iq3_xxs blk.35.ffn_up_exps.weight iq3_xxs +blk.35.ffn_up_shexp.weight iq3_xxs blk.36.attn_qkv.weight q4_K -blk.36.ffn_gate_shexp.weight iq3_xxs -blk.36.ffn_up_shexp.weight iq3_xxs blk.36.ffn_gate_exps.weight iq3_xxs +blk.36.ffn_gate_shexp.weight iq3_xxs blk.36.ffn_up_exps.weight iq3_xxs +blk.36.ffn_up_shexp.weight iq3_xxs blk.37.attn_qkv.weight q4_K -blk.37.ffn_gate_shexp.weight iq3_xxs -blk.37.ffn_up_shexp.weight iq3_xxs blk.37.ffn_gate_exps.weight iq3_xxs +blk.37.ffn_gate_shexp.weight iq3_xxs blk.37.ffn_up_exps.weight iq3_xxs +blk.37.ffn_up_shexp.weight iq3_xxs blk.38.attn_qkv.weight q4_K -blk.38.ffn_gate_shexp.weight iq3_xxs -blk.38.ffn_up_shexp.weight iq3_xxs blk.38.ffn_gate_exps.weight iq3_xxs +blk.38.ffn_gate_shexp.weight iq3_xxs blk.38.ffn_up_exps.weight iq3_xxs -blk.39.ffn_gate_shexp.weight iq3_xxs -blk.39.ffn_up_shexp.weight iq3_xxs +blk.38.ffn_up_shexp.weight iq3_xxs blk.39.attn_k.weight iq3_xxs blk.39.attn_q.weight iq3_xxs blk.39.attn_v.weight q4_K blk.39.ffn_gate_exps.weight iq3_xxs +blk.39.ffn_gate_shexp.weight iq3_xxs blk.39.ffn_up_exps.weight iq3_xxs +blk.39.ffn_up_shexp.weight iq3_xxs blk.40.attn_qkv.weight q4_K blk.40.ffn_gate_exps.weight iq3_xxs -blk.40.ffn_up_exps.weight iq3_xxs blk.40.ffn_gate_shexp.weight iq3_xxs +blk.40.ffn_up_exps.weight iq3_xxs blk.40.ffn_up_shexp.weight iq3_xxs blk.41.attn_qkv.weight q4_K -blk.41.ffn_gate_shexp.weight iq3_xxs -blk.41.ffn_up_shexp.weight iq3_xxs blk.41.ffn_gate_exps.weight iq3_xxs +blk.41.ffn_gate_shexp.weight iq3_xxs blk.41.ffn_up_exps.weight iq3_xxs +blk.41.ffn_up_shexp.weight iq3_xxs blk.42.attn_qkv.weight q4_K blk.43.attn_k.weight iq3_xxs blk.43.attn_q.weight iq3_xxs @@ -1069,203 +1070,203 @@ blk.46.attn_qkv.weight q4_K blk.47.attn_k.weight iq3_xxs blk.47.attn_q.weight iq3_xxs blk.47.attn_v.weight q4_K -output.weight q6_K [IQ3_XXS] iq3_xxs +output.weight q5_K token_embd.weight iq3_s blk.0.attn_qkv.weight q4_K blk.0.ffn_down_exps.weight q4_K blk.0.ffn_down_shexp.weight q4_K blk.1.attn_qkv.weight q4_K -blk.1.ffn_down_shexp.weight q4_K blk.1.ffn_down_exps.weight q4_K +blk.1.ffn_down_shexp.weight q4_K blk.2.attn_qkv.weight q4_K -blk.2.ffn_down_shexp.weight q4_K blk.2.ffn_down_exps.weight q4_K -blk.3.ffn_down_shexp.weight q4_K +blk.2.ffn_down_shexp.weight q4_K blk.3.attn_k.weight iq2_s blk.3.attn_output.weight iq3_s blk.3.attn_q.weight iq2_s blk.3.attn_v.weight q4_K blk.3.ffn_down_exps.weight q4_K +blk.3.ffn_down_shexp.weight q4_K blk.4.attn_qkv.weight q4_K -blk.4.ffn_down_shexp.weight q4_K blk.4.ffn_down_exps.weight q4_K +blk.4.ffn_down_shexp.weight q4_K blk.5.attn_qkv.weight q4_K -blk.5.ffn_down_shexp.weight q4_K blk.5.ffn_down_exps.weight q4_K +blk.5.ffn_down_shexp.weight q4_K blk.6.attn_qkv.weight q4_K blk.6.ffn_down_exps.weight q3_K blk.6.ffn_down_shexp.weight q3_K -blk.7.ffn_down_shexp.weight q3_K blk.7.attn_k.weight iq2_s blk.7.attn_output.weight iq3_s blk.7.attn_q.weight iq2_s blk.7.attn_v.weight q4_K blk.7.ffn_down_exps.weight q3_K +blk.7.ffn_down_shexp.weight q3_K blk.8.attn_qkv.weight q4_K -blk.8.ffn_down_shexp.weight q3_K blk.8.ffn_down_exps.weight q3_K +blk.8.ffn_down_shexp.weight q3_K blk.9.attn_qkv.weight q4_K +blk.9.ffn_down_exps.weight q3_K blk.9.ffn_down_shexp.weight q3_K blk.10.attn_qkv.weight q4_K -blk.10.ffn_down_shexp.weight q3_K -blk.9.ffn_down_exps.weight q3_K blk.10.ffn_down_exps.weight q3_K -blk.11.ffn_down_shexp.weight q3_K +blk.10.ffn_down_shexp.weight q3_K blk.11.attn_k.weight iq2_s blk.11.attn_output.weight iq3_s blk.11.attn_q.weight iq2_s blk.11.attn_v.weight q4_K blk.11.ffn_down_exps.weight q3_K +blk.11.ffn_down_shexp.weight q3_K blk.12.attn_qkv.weight q4_K blk.12.ffn_down_exps.weight q3_K blk.12.ffn_down_shexp.weight q3_K blk.13.attn_qkv.weight q4_K -blk.13.ffn_down_shexp.weight q3_K blk.13.ffn_down_exps.weight q3_K +blk.13.ffn_down_shexp.weight q3_K blk.14.attn_qkv.weight q4_K -blk.14.ffn_down_shexp.weight q3_K blk.14.ffn_down_exps.weight q3_K -blk.15.ffn_down_shexp.weight q3_K +blk.14.ffn_down_shexp.weight q3_K blk.15.attn_k.weight iq2_s blk.15.attn_output.weight iq3_s blk.15.attn_q.weight iq2_s blk.15.attn_v.weight q4_K blk.15.ffn_down_exps.weight q3_K +blk.15.ffn_down_shexp.weight q3_K blk.16.attn_qkv.weight q4_K -blk.16.ffn_down_shexp.weight q3_K blk.16.ffn_down_exps.weight q3_K +blk.16.ffn_down_shexp.weight q3_K blk.17.attn_qkv.weight q4_K blk.17.ffn_down_exps.weight q3_K blk.17.ffn_down_shexp.weight q3_K blk.18.attn_qkv.weight q4_K -blk.18.ffn_down_shexp.weight q3_K blk.18.ffn_down_exps.weight q3_K -blk.19.ffn_down_shexp.weight q3_K +blk.18.ffn_down_shexp.weight q3_K blk.19.attn_k.weight iq2_s blk.19.attn_output.weight iq3_s blk.19.attn_q.weight iq2_s blk.19.attn_v.weight q4_K blk.19.ffn_down_exps.weight q3_K +blk.19.ffn_down_shexp.weight q3_K blk.20.attn_qkv.weight q4_K -blk.20.ffn_down_shexp.weight q3_K blk.20.ffn_down_exps.weight q3_K +blk.20.ffn_down_shexp.weight q3_K blk.21.attn_qkv.weight q4_K -blk.21.ffn_down_shexp.weight q3_K blk.21.ffn_down_exps.weight q3_K +blk.21.ffn_down_shexp.weight q3_K blk.22.attn_qkv.weight q4_K -blk.22.ffn_down_shexp.weight q3_K blk.22.ffn_down_exps.weight q3_K -blk.23.ffn_down_exps.weight q3_K -blk.23.ffn_down_shexp.weight q3_K +blk.22.ffn_down_shexp.weight q3_K blk.23.attn_k.weight iq2_s blk.23.attn_output.weight iq3_s blk.23.attn_q.weight iq2_s blk.23.attn_v.weight q4_K +blk.23.ffn_down_exps.weight q3_K +blk.23.ffn_down_shexp.weight q3_K blk.24.attn_qkv.weight q4_K -blk.24.ffn_down_shexp.weight q3_K blk.24.ffn_down_exps.weight q3_K +blk.24.ffn_down_shexp.weight q3_K blk.25.attn_qkv.weight q4_K -blk.25.ffn_down_shexp.weight q3_K blk.25.ffn_down_exps.weight q3_K +blk.25.ffn_down_shexp.weight q3_K blk.26.attn_qkv.weight q4_K -blk.26.ffn_down_shexp.weight q3_K blk.26.ffn_down_exps.weight q3_K -blk.27.ffn_down_shexp.weight q3_K +blk.26.ffn_down_shexp.weight q3_K blk.27.attn_k.weight iq2_s blk.27.attn_output.weight iq3_s blk.27.attn_q.weight iq2_s blk.27.attn_v.weight q4_K blk.27.ffn_down_exps.weight q3_K +blk.27.ffn_down_shexp.weight q3_K blk.28.attn_qkv.weight q4_K -blk.28.ffn_down_shexp.weight q3_K blk.28.ffn_down_exps.weight q3_K +blk.28.ffn_down_shexp.weight q3_K blk.29.attn_qkv.weight q4_K blk.29.ffn_down_exps.weight q3_K blk.29.ffn_down_shexp.weight q3_K blk.30.attn_qkv.weight q4_K -blk.30.ffn_down_shexp.weight q3_K blk.30.ffn_down_exps.weight q3_K -blk.31.ffn_down_shexp.weight q3_K +blk.30.ffn_down_shexp.weight q3_K blk.31.attn_k.weight iq2_s blk.31.attn_output.weight iq3_s blk.31.attn_q.weight iq2_s blk.31.attn_v.weight q4_K blk.31.ffn_down_exps.weight q3_K +blk.31.ffn_down_shexp.weight q3_K blk.32.attn_qkv.weight q4_K -blk.32.ffn_down_shexp.weight q3_K blk.32.ffn_down_exps.weight q3_K +blk.32.ffn_down_shexp.weight q3_K blk.33.attn_qkv.weight q4_K -blk.33.ffn_down_shexp.weight q3_K blk.33.ffn_down_exps.weight q3_K +blk.33.ffn_down_shexp.weight q3_K blk.34.attn_qkv.weight q4_K blk.34.ffn_down_exps.weight q3_K blk.34.ffn_down_shexp.weight q3_K -blk.35.ffn_down_shexp.weight q3_K blk.35.attn_k.weight iq2_s blk.35.attn_output.weight iq3_s blk.35.attn_q.weight iq2_s blk.35.attn_v.weight q4_K blk.35.ffn_down_exps.weight q3_K +blk.35.ffn_down_shexp.weight q3_K blk.36.attn_qkv.weight q4_K -blk.36.ffn_down_shexp.weight q3_K blk.36.ffn_down_exps.weight q3_K +blk.36.ffn_down_shexp.weight q3_K blk.37.attn_qkv.weight q4_K -blk.37.ffn_down_shexp.weight q3_K blk.37.ffn_down_exps.weight q3_K +blk.37.ffn_down_shexp.weight q3_K blk.38.attn_qkv.weight q4_K -blk.38.ffn_down_shexp.weight q3_K blk.38.ffn_down_exps.weight q3_K -blk.39.ffn_down_shexp.weight q3_K +blk.38.ffn_down_shexp.weight q3_K blk.39.attn_k.weight iq2_s blk.39.attn_output.weight iq3_s blk.39.attn_q.weight iq2_s blk.39.attn_v.weight q4_K blk.39.ffn_down_exps.weight q3_K +blk.39.ffn_down_shexp.weight q3_K blk.40.attn_qkv.weight q4_K blk.40.ffn_down_exps.weight q3_K blk.40.ffn_down_shexp.weight q3_K blk.41.attn_qkv.weight q4_K -blk.41.ffn_down_shexp.weight q3_K blk.41.ffn_down_exps.weight q3_K +blk.41.ffn_down_shexp.weight q3_K blk.42.attn_qkv.weight q4_K -blk.42.ffn_down_shexp.weight q3_K blk.42.ffn_down_exps.weight q3_K -blk.43.ffn_down_shexp.weight q3_K +blk.42.ffn_down_shexp.weight q3_K blk.43.attn_k.weight iq2_s blk.43.attn_output.weight iq3_s blk.43.attn_q.weight iq2_s blk.43.attn_v.weight q4_K blk.43.ffn_down_exps.weight q3_K +blk.43.ffn_down_shexp.weight q3_K blk.44.attn_qkv.weight q4_K -blk.44.ffn_down_shexp.weight q3_K blk.44.ffn_down_exps.weight q3_K +blk.44.ffn_down_shexp.weight q3_K blk.45.attn_qkv.weight q4_K -blk.45.ffn_down_shexp.weight q3_K blk.45.ffn_down_exps.weight q3_K +blk.45.ffn_down_shexp.weight q3_K blk.46.attn_qkv.weight q4_K blk.46.ffn_down_exps.weight q3_K blk.46.ffn_down_shexp.weight q3_K -blk.47.ffn_down_shexp.weight q3_K blk.47.attn_k.weight iq2_s blk.47.attn_output.weight iq3_s blk.47.attn_q.weight iq2_s blk.47.attn_v.weight q4_K -output.weight q5_K blk.47.ffn_down_exps.weight q3_K +blk.47.ffn_down_shexp.weight q3_K [IQ1_S] iq1_s +output.weight q5_K token_embd.weight q2_K blk.0.attn_qkv.weight q4_K blk.0.ffn_down_exps.weight q2_K blk.0.ffn_down_shexp.weight q2_K blk.1.attn_qkv.weight q4_K -blk.1.ffn_down_shexp.weight q2_K blk.1.ffn_down_exps.weight q2_K +blk.1.ffn_down_shexp.weight q2_K blk.2.attn_qkv.weight q4_K -blk.2.ffn_down_shexp.weight q2_K blk.2.ffn_down_exps.weight q2_K +blk.2.ffn_down_shexp.weight q2_K blk.3.attn_output.weight iq2_xxs blk.3.attn_v.weight q4_K blk.4.attn_qkv.weight q4_K @@ -1323,27 +1324,27 @@ blk.45.attn_qkv.weight q4_K blk.46.attn_qkv.weight q4_K blk.47.attn_output.weight iq2_xxs blk.47.attn_v.weight q4_K -output.weight q5_K [IQ4_NL] iq4_nl +output.weight q6_K blk.0.attn_qkv.weight q5_K blk.0.ffn_down_exps.weight q5_K blk.0.ffn_down_shexp.weight q5_K blk.1.attn_qkv.weight q5_K -blk.1.ffn_down_shexp.weight q5_K blk.1.ffn_down_exps.weight q5_K +blk.1.ffn_down_shexp.weight q5_K blk.2.attn_qkv.weight q5_K -blk.2.ffn_down_shexp.weight q5_K blk.2.ffn_down_exps.weight q5_K -blk.3.ffn_down_shexp.weight q5_K +blk.2.ffn_down_shexp.weight q5_K blk.3.attn_v.weight q5_K blk.3.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q5_K blk.4.attn_qkv.weight q5_K -blk.4.ffn_down_shexp.weight q5_K blk.4.ffn_down_exps.weight q5_K +blk.4.ffn_down_shexp.weight q5_K blk.5.attn_qkv.weight q5_K -blk.5.ffn_down_shexp.weight q5_K blk.5.ffn_down_exps.weight q5_K +blk.5.ffn_down_shexp.weight q5_K blk.6.attn_qkv.weight q5_K blk.7.attn_v.weight q5_K blk.8.attn_qkv.weight q5_K @@ -1386,9 +1387,9 @@ blk.44.attn_qkv.weight q5_K blk.45.attn_qkv.weight q5_K blk.46.attn_qkv.weight q5_K blk.47.attn_v.weight q5_K -output.weight q6_K [IQ3_S] iq3_s +output.weight q6_K blk.0.attn_qkv.weight q4_K blk.1.attn_qkv.weight q4_K blk.2.attn_qkv.weight q4_K @@ -1437,28 +1438,28 @@ blk.44.attn_qkv.weight q4_K blk.45.attn_qkv.weight q4_K blk.46.attn_qkv.weight q4_K blk.47.attn_v.weight q4_K -output.weight q6_K [IQ3_M] iq3_s +output.weight q6_K blk.0.attn_qkv.weight q4_K blk.0.ffn_down_exps.weight q4_K blk.0.ffn_down_shexp.weight q4_K blk.1.attn_qkv.weight q4_K -blk.1.ffn_down_shexp.weight q4_K blk.1.ffn_down_exps.weight q4_K +blk.1.ffn_down_shexp.weight q4_K blk.2.attn_qkv.weight q4_K -blk.2.ffn_down_shexp.weight q4_K blk.2.ffn_down_exps.weight q4_K -blk.3.ffn_down_shexp.weight q4_K +blk.2.ffn_down_shexp.weight q4_K blk.3.attn_output.weight q4_K blk.3.attn_v.weight q4_K blk.3.ffn_down_exps.weight q4_K +blk.3.ffn_down_shexp.weight q4_K blk.4.attn_qkv.weight q4_K -blk.4.ffn_down_shexp.weight q4_K blk.4.ffn_down_exps.weight q4_K +blk.4.ffn_down_shexp.weight q4_K blk.5.attn_qkv.weight q4_K -blk.5.ffn_down_shexp.weight q4_K blk.5.ffn_down_exps.weight q4_K +blk.5.ffn_down_shexp.weight q4_K blk.6.attn_qkv.weight q4_K blk.7.attn_output.weight q4_K blk.7.attn_v.weight q4_K @@ -1512,19 +1513,19 @@ blk.45.attn_qkv.weight q4_K blk.46.attn_qkv.weight q4_K blk.47.attn_output.weight q4_K blk.47.attn_v.weight q4_K -output.weight q6_K [IQ2_S] iq2_xs +output.weight q5_K token_embd.weight iq3_s blk.0.attn_qkv.weight q4_K blk.0.ffn_down_exps.weight iq3_s blk.0.ffn_down_shexp.weight iq3_s blk.1.attn_qkv.weight q4_K -blk.1.ffn_down_shexp.weight iq3_s blk.1.ffn_down_exps.weight iq3_s +blk.1.ffn_down_shexp.weight iq3_s blk.2.attn_qkv.weight q4_K -blk.2.ffn_down_shexp.weight iq3_s blk.2.ffn_down_exps.weight iq3_s +blk.2.ffn_down_shexp.weight iq3_s blk.3.attn_output.weight iq3_s blk.3.attn_v.weight q4_K blk.4.attn_qkv.weight q4_K @@ -1582,19 +1583,19 @@ blk.45.attn_qkv.weight q4_K blk.46.attn_qkv.weight q4_K blk.47.attn_output.weight iq3_s blk.47.attn_v.weight q4_K -output.weight q5_K [IQ2_M] iq2_s +output.weight q5_K token_embd.weight iq3_s blk.0.attn_qkv.weight q4_K blk.0.ffn_down_exps.weight iq3_s blk.0.ffn_down_shexp.weight iq3_s blk.1.attn_qkv.weight q4_K -blk.1.ffn_down_shexp.weight iq3_s blk.1.ffn_down_exps.weight iq3_s +blk.1.ffn_down_shexp.weight iq3_s blk.2.attn_qkv.weight q4_K -blk.2.ffn_down_shexp.weight iq3_s blk.2.ffn_down_exps.weight iq3_s +blk.2.ffn_down_shexp.weight iq3_s blk.3.attn_output.weight iq3_s blk.3.attn_v.weight q4_K blk.4.attn_qkv.weight q4_K @@ -1652,27 +1653,27 @@ blk.45.attn_qkv.weight q4_K blk.46.attn_qkv.weight q4_K blk.47.attn_output.weight iq3_s blk.47.attn_v.weight q4_K -output.weight q5_K [IQ4_XS] iq4_xs +output.weight q6_K blk.0.attn_qkv.weight q5_K blk.0.ffn_down_exps.weight q5_K blk.0.ffn_down_shexp.weight q5_K blk.1.attn_qkv.weight q5_K -blk.1.ffn_down_shexp.weight q5_K blk.1.ffn_down_exps.weight q5_K +blk.1.ffn_down_shexp.weight q5_K blk.2.attn_qkv.weight q5_K -blk.2.ffn_down_shexp.weight q5_K blk.2.ffn_down_exps.weight q5_K -blk.3.ffn_down_shexp.weight q5_K +blk.2.ffn_down_shexp.weight q5_K blk.3.attn_v.weight q5_K blk.3.ffn_down_exps.weight q5_K +blk.3.ffn_down_shexp.weight q5_K blk.4.attn_qkv.weight q5_K -blk.4.ffn_down_shexp.weight q5_K blk.4.ffn_down_exps.weight q5_K +blk.4.ffn_down_shexp.weight q5_K blk.5.attn_qkv.weight q5_K -blk.5.ffn_down_shexp.weight q5_K blk.5.ffn_down_exps.weight q5_K +blk.5.ffn_down_shexp.weight q5_K blk.6.attn_qkv.weight q5_K blk.7.attn_v.weight q5_K blk.8.attn_qkv.weight q5_K @@ -1715,19 +1716,19 @@ blk.44.attn_qkv.weight q5_K blk.45.attn_qkv.weight q5_K blk.46.attn_qkv.weight q5_K blk.47.attn_v.weight q5_K -output.weight q6_K [IQ1_M] iq1_m +output.weight q5_K token_embd.weight q2_K blk.0.attn_qkv.weight q4_K blk.0.ffn_down_exps.weight q2_K blk.0.ffn_down_shexp.weight q2_K blk.1.attn_qkv.weight q4_K -blk.1.ffn_down_shexp.weight q2_K blk.1.ffn_down_exps.weight q2_K +blk.1.ffn_down_shexp.weight q2_K blk.2.attn_qkv.weight q4_K -blk.2.ffn_down_shexp.weight q2_K blk.2.ffn_down_exps.weight q2_K +blk.2.ffn_down_shexp.weight q2_K blk.3.attn_output.weight iq2_xxs blk.3.attn_v.weight q4_K blk.4.attn_qkv.weight q4_K @@ -1785,354 +1786,353 @@ blk.45.attn_qkv.weight q4_K blk.46.attn_qkv.weight q4_K blk.47.attn_output.weight iq2_xxs blk.47.attn_v.weight q4_K -output.weight q5_K [BF16] bf16 [TQ1_0] tq1_0 -token_embd.weight q4_K output.weight q6_K +token_embd.weight q4_K [TQ2_0] tq2_0 -token_embd.weight q4_K output.weight q6_K +token_embd.weight q4_K [MXFP4_MOE] mxfp4 +output.weight q8_0 token_embd.weight q8_0 -blk.0.ssm_ba.weight q8_0 -blk.0.attn_qkv.weight q8_0 blk.0.attn_gate.weight q8_0 -blk.0.ssm_out.weight q8_0 +blk.0.attn_qkv.weight q8_0 blk.0.ffn_down_shexp.weight q8_0 blk.0.ffn_gate_shexp.weight q8_0 blk.0.ffn_up_shexp.weight q8_0 -blk.1.ssm_ba.weight q8_0 -blk.1.attn_qkv.weight q8_0 +blk.0.ssm_ba.weight q8_0 +blk.0.ssm_out.weight q8_0 blk.1.attn_gate.weight q8_0 -blk.1.ssm_out.weight q8_0 +blk.1.attn_qkv.weight q8_0 blk.1.ffn_down_shexp.weight q8_0 blk.1.ffn_gate_shexp.weight q8_0 blk.1.ffn_up_shexp.weight q8_0 -blk.2.ssm_ba.weight q8_0 -blk.2.attn_qkv.weight q8_0 +blk.1.ssm_ba.weight q8_0 +blk.1.ssm_out.weight q8_0 blk.2.attn_gate.weight q8_0 -blk.2.ssm_out.weight q8_0 +blk.2.attn_qkv.weight q8_0 blk.2.ffn_down_shexp.weight q8_0 blk.2.ffn_gate_shexp.weight q8_0 blk.2.ffn_up_shexp.weight q8_0 -blk.3.ffn_down_shexp.weight q8_0 -blk.3.ffn_gate_shexp.weight q8_0 -blk.3.ffn_up_shexp.weight q8_0 +blk.2.ssm_ba.weight q8_0 +blk.2.ssm_out.weight q8_0 blk.3.attn_k.weight q8_0 blk.3.attn_output.weight q8_0 blk.3.attn_q.weight q8_0 blk.3.attn_v.weight q8_0 -blk.4.ssm_ba.weight q8_0 -blk.4.attn_qkv.weight q8_0 +blk.3.ffn_down_shexp.weight q8_0 +blk.3.ffn_gate_shexp.weight q8_0 +blk.3.ffn_up_shexp.weight q8_0 blk.4.attn_gate.weight q8_0 -blk.4.ssm_out.weight q8_0 +blk.4.attn_qkv.weight q8_0 blk.4.ffn_down_shexp.weight q8_0 blk.4.ffn_gate_shexp.weight q8_0 blk.4.ffn_up_shexp.weight q8_0 -blk.5.ssm_ba.weight q8_0 -blk.5.attn_qkv.weight q8_0 +blk.4.ssm_ba.weight q8_0 +blk.4.ssm_out.weight q8_0 blk.5.attn_gate.weight q8_0 -blk.5.ssm_out.weight q8_0 +blk.5.attn_qkv.weight q8_0 blk.5.ffn_down_shexp.weight q8_0 blk.5.ffn_gate_shexp.weight q8_0 blk.5.ffn_up_shexp.weight q8_0 -blk.6.ssm_ba.weight q8_0 -blk.6.attn_qkv.weight q8_0 +blk.5.ssm_ba.weight q8_0 +blk.5.ssm_out.weight q8_0 blk.6.attn_gate.weight q8_0 -blk.6.ssm_out.weight q8_0 +blk.6.attn_qkv.weight q8_0 blk.6.ffn_down_shexp.weight q8_0 blk.6.ffn_gate_shexp.weight q8_0 blk.6.ffn_up_shexp.weight q8_0 -blk.7.ffn_down_shexp.weight q8_0 -blk.7.ffn_gate_shexp.weight q8_0 -blk.7.ffn_up_shexp.weight q8_0 +blk.6.ssm_ba.weight q8_0 +blk.6.ssm_out.weight q8_0 blk.7.attn_k.weight q8_0 blk.7.attn_output.weight q8_0 blk.7.attn_q.weight q8_0 blk.7.attn_v.weight q8_0 -blk.8.ssm_ba.weight q8_0 -blk.8.attn_qkv.weight q8_0 +blk.7.ffn_down_shexp.weight q8_0 +blk.7.ffn_gate_shexp.weight q8_0 +blk.7.ffn_up_shexp.weight q8_0 blk.8.attn_gate.weight q8_0 -blk.8.ssm_out.weight q8_0 +blk.8.attn_qkv.weight q8_0 blk.8.ffn_down_shexp.weight q8_0 blk.8.ffn_gate_shexp.weight q8_0 blk.8.ffn_up_shexp.weight q8_0 -blk.9.ssm_ba.weight q8_0 -blk.9.attn_qkv.weight q8_0 +blk.8.ssm_ba.weight q8_0 +blk.8.ssm_out.weight q8_0 blk.9.attn_gate.weight q8_0 -blk.9.ssm_out.weight q8_0 +blk.9.attn_qkv.weight q8_0 blk.9.ffn_down_shexp.weight q8_0 blk.9.ffn_gate_shexp.weight q8_0 blk.9.ffn_up_shexp.weight q8_0 -blk.10.ssm_ba.weight q8_0 -blk.10.attn_qkv.weight q8_0 +blk.9.ssm_ba.weight q8_0 +blk.9.ssm_out.weight q8_0 blk.10.attn_gate.weight q8_0 -blk.10.ssm_out.weight q8_0 +blk.10.attn_qkv.weight q8_0 blk.10.ffn_down_shexp.weight q8_0 blk.10.ffn_gate_shexp.weight q8_0 blk.10.ffn_up_shexp.weight q8_0 -blk.11.ffn_down_shexp.weight q8_0 -blk.11.ffn_gate_shexp.weight q8_0 -blk.11.ffn_up_shexp.weight q8_0 +blk.10.ssm_ba.weight q8_0 +blk.10.ssm_out.weight q8_0 blk.11.attn_k.weight q8_0 blk.11.attn_output.weight q8_0 blk.11.attn_q.weight q8_0 blk.11.attn_v.weight q8_0 -blk.12.ssm_ba.weight q8_0 -blk.12.attn_qkv.weight q8_0 +blk.11.ffn_down_shexp.weight q8_0 +blk.11.ffn_gate_shexp.weight q8_0 +blk.11.ffn_up_shexp.weight q8_0 blk.12.attn_gate.weight q8_0 -blk.12.ssm_out.weight q8_0 +blk.12.attn_qkv.weight q8_0 blk.12.ffn_down_shexp.weight q8_0 blk.12.ffn_gate_shexp.weight q8_0 blk.12.ffn_up_shexp.weight q8_0 -blk.13.ssm_ba.weight q8_0 -blk.13.attn_qkv.weight q8_0 +blk.12.ssm_ba.weight q8_0 +blk.12.ssm_out.weight q8_0 blk.13.attn_gate.weight q8_0 -blk.13.ssm_out.weight q8_0 +blk.13.attn_qkv.weight q8_0 blk.13.ffn_down_shexp.weight q8_0 blk.13.ffn_gate_shexp.weight q8_0 blk.13.ffn_up_shexp.weight q8_0 -blk.14.ssm_ba.weight q8_0 -blk.14.attn_qkv.weight q8_0 +blk.13.ssm_ba.weight q8_0 +blk.13.ssm_out.weight q8_0 blk.14.attn_gate.weight q8_0 -blk.14.ssm_out.weight q8_0 +blk.14.attn_qkv.weight q8_0 blk.14.ffn_down_shexp.weight q8_0 blk.14.ffn_gate_shexp.weight q8_0 blk.14.ffn_up_shexp.weight q8_0 -blk.15.ffn_down_shexp.weight q8_0 -blk.15.ffn_gate_shexp.weight q8_0 -blk.15.ffn_up_shexp.weight q8_0 +blk.14.ssm_ba.weight q8_0 +blk.14.ssm_out.weight q8_0 blk.15.attn_k.weight q8_0 blk.15.attn_output.weight q8_0 blk.15.attn_q.weight q8_0 blk.15.attn_v.weight q8_0 -blk.16.ssm_ba.weight q8_0 -blk.16.attn_qkv.weight q8_0 +blk.15.ffn_down_shexp.weight q8_0 +blk.15.ffn_gate_shexp.weight q8_0 +blk.15.ffn_up_shexp.weight q8_0 blk.16.attn_gate.weight q8_0 -blk.16.ssm_out.weight q8_0 +blk.16.attn_qkv.weight q8_0 blk.16.ffn_down_shexp.weight q8_0 blk.16.ffn_gate_shexp.weight q8_0 blk.16.ffn_up_shexp.weight q8_0 -blk.17.ssm_ba.weight q8_0 -blk.17.attn_qkv.weight q8_0 +blk.16.ssm_ba.weight q8_0 +blk.16.ssm_out.weight q8_0 blk.17.attn_gate.weight q8_0 -blk.17.ssm_out.weight q8_0 +blk.17.attn_qkv.weight q8_0 blk.17.ffn_down_shexp.weight q8_0 blk.17.ffn_gate_shexp.weight q8_0 blk.17.ffn_up_shexp.weight q8_0 -blk.18.ssm_ba.weight q8_0 -blk.18.attn_qkv.weight q8_0 +blk.17.ssm_ba.weight q8_0 +blk.17.ssm_out.weight q8_0 blk.18.attn_gate.weight q8_0 -blk.18.ssm_out.weight q8_0 +blk.18.attn_qkv.weight q8_0 blk.18.ffn_down_shexp.weight q8_0 blk.18.ffn_gate_shexp.weight q8_0 blk.18.ffn_up_shexp.weight q8_0 -blk.19.ffn_down_shexp.weight q8_0 -blk.19.ffn_gate_shexp.weight q8_0 -blk.19.ffn_up_shexp.weight q8_0 +blk.18.ssm_ba.weight q8_0 +blk.18.ssm_out.weight q8_0 blk.19.attn_k.weight q8_0 blk.19.attn_output.weight q8_0 blk.19.attn_q.weight q8_0 blk.19.attn_v.weight q8_0 -blk.20.ssm_ba.weight q8_0 -blk.20.attn_qkv.weight q8_0 +blk.19.ffn_down_shexp.weight q8_0 +blk.19.ffn_gate_shexp.weight q8_0 +blk.19.ffn_up_shexp.weight q8_0 blk.20.attn_gate.weight q8_0 -blk.20.ssm_out.weight q8_0 +blk.20.attn_qkv.weight q8_0 blk.20.ffn_down_shexp.weight q8_0 blk.20.ffn_gate_shexp.weight q8_0 blk.20.ffn_up_shexp.weight q8_0 -blk.21.ssm_ba.weight q8_0 -blk.21.attn_qkv.weight q8_0 +blk.20.ssm_ba.weight q8_0 +blk.20.ssm_out.weight q8_0 blk.21.attn_gate.weight q8_0 -blk.21.ssm_out.weight q8_0 +blk.21.attn_qkv.weight q8_0 blk.21.ffn_down_shexp.weight q8_0 blk.21.ffn_gate_shexp.weight q8_0 blk.21.ffn_up_shexp.weight q8_0 -blk.22.ssm_ba.weight q8_0 -blk.22.attn_qkv.weight q8_0 +blk.21.ssm_ba.weight q8_0 +blk.21.ssm_out.weight q8_0 blk.22.attn_gate.weight q8_0 -blk.22.ssm_out.weight q8_0 +blk.22.attn_qkv.weight q8_0 blk.22.ffn_down_shexp.weight q8_0 blk.22.ffn_gate_shexp.weight q8_0 blk.22.ffn_up_shexp.weight q8_0 -blk.23.ffn_down_shexp.weight q8_0 -blk.23.ffn_gate_shexp.weight q8_0 -blk.23.ffn_up_shexp.weight q8_0 +blk.22.ssm_ba.weight q8_0 +blk.22.ssm_out.weight q8_0 blk.23.attn_k.weight q8_0 blk.23.attn_output.weight q8_0 blk.23.attn_q.weight q8_0 blk.23.attn_v.weight q8_0 -blk.24.ssm_ba.weight q8_0 -blk.24.attn_qkv.weight q8_0 +blk.23.ffn_down_shexp.weight q8_0 +blk.23.ffn_gate_shexp.weight q8_0 +blk.23.ffn_up_shexp.weight q8_0 blk.24.attn_gate.weight q8_0 -blk.24.ssm_out.weight q8_0 +blk.24.attn_qkv.weight q8_0 blk.24.ffn_down_shexp.weight q8_0 blk.24.ffn_gate_shexp.weight q8_0 blk.24.ffn_up_shexp.weight q8_0 -blk.25.ssm_ba.weight q8_0 -blk.25.attn_qkv.weight q8_0 +blk.24.ssm_ba.weight q8_0 +blk.24.ssm_out.weight q8_0 blk.25.attn_gate.weight q8_0 -blk.25.ssm_out.weight q8_0 +blk.25.attn_qkv.weight q8_0 blk.25.ffn_down_shexp.weight q8_0 blk.25.ffn_gate_shexp.weight q8_0 blk.25.ffn_up_shexp.weight q8_0 -blk.26.ssm_ba.weight q8_0 -blk.26.attn_qkv.weight q8_0 +blk.25.ssm_ba.weight q8_0 +blk.25.ssm_out.weight q8_0 blk.26.attn_gate.weight q8_0 -blk.26.ssm_out.weight q8_0 +blk.26.attn_qkv.weight q8_0 blk.26.ffn_down_shexp.weight q8_0 blk.26.ffn_gate_shexp.weight q8_0 blk.26.ffn_up_shexp.weight q8_0 -blk.27.ffn_down_shexp.weight q8_0 -blk.27.ffn_gate_shexp.weight q8_0 -blk.27.ffn_up_shexp.weight q8_0 +blk.26.ssm_ba.weight q8_0 +blk.26.ssm_out.weight q8_0 blk.27.attn_k.weight q8_0 blk.27.attn_output.weight q8_0 blk.27.attn_q.weight q8_0 blk.27.attn_v.weight q8_0 -blk.28.ssm_ba.weight q8_0 -blk.28.attn_qkv.weight q8_0 +blk.27.ffn_down_shexp.weight q8_0 +blk.27.ffn_gate_shexp.weight q8_0 +blk.27.ffn_up_shexp.weight q8_0 blk.28.attn_gate.weight q8_0 -blk.28.ssm_out.weight q8_0 +blk.28.attn_qkv.weight q8_0 blk.28.ffn_down_shexp.weight q8_0 blk.28.ffn_gate_shexp.weight q8_0 blk.28.ffn_up_shexp.weight q8_0 -blk.29.ssm_ba.weight q8_0 -blk.29.attn_qkv.weight q8_0 +blk.28.ssm_ba.weight q8_0 +blk.28.ssm_out.weight q8_0 blk.29.attn_gate.weight q8_0 -blk.29.ssm_out.weight q8_0 +blk.29.attn_qkv.weight q8_0 blk.29.ffn_down_shexp.weight q8_0 blk.29.ffn_gate_shexp.weight q8_0 blk.29.ffn_up_shexp.weight q8_0 -blk.30.ssm_ba.weight q8_0 -blk.30.attn_qkv.weight q8_0 +blk.29.ssm_ba.weight q8_0 +blk.29.ssm_out.weight q8_0 blk.30.attn_gate.weight q8_0 -blk.30.ssm_out.weight q8_0 +blk.30.attn_qkv.weight q8_0 blk.30.ffn_down_shexp.weight q8_0 blk.30.ffn_gate_shexp.weight q8_0 blk.30.ffn_up_shexp.weight q8_0 -blk.31.ffn_down_shexp.weight q8_0 -blk.31.ffn_gate_shexp.weight q8_0 -blk.31.ffn_up_shexp.weight q8_0 +blk.30.ssm_ba.weight q8_0 +blk.30.ssm_out.weight q8_0 blk.31.attn_k.weight q8_0 blk.31.attn_output.weight q8_0 blk.31.attn_q.weight q8_0 blk.31.attn_v.weight q8_0 -blk.32.ssm_ba.weight q8_0 -blk.32.attn_qkv.weight q8_0 +blk.31.ffn_down_shexp.weight q8_0 +blk.31.ffn_gate_shexp.weight q8_0 +blk.31.ffn_up_shexp.weight q8_0 blk.32.attn_gate.weight q8_0 -blk.32.ssm_out.weight q8_0 +blk.32.attn_qkv.weight q8_0 blk.32.ffn_down_shexp.weight q8_0 blk.32.ffn_gate_shexp.weight q8_0 blk.32.ffn_up_shexp.weight q8_0 -blk.33.ssm_ba.weight q8_0 -blk.33.attn_qkv.weight q8_0 +blk.32.ssm_ba.weight q8_0 +blk.32.ssm_out.weight q8_0 blk.33.attn_gate.weight q8_0 -blk.33.ssm_out.weight q8_0 +blk.33.attn_qkv.weight q8_0 blk.33.ffn_down_shexp.weight q8_0 blk.33.ffn_gate_shexp.weight q8_0 blk.33.ffn_up_shexp.weight q8_0 -blk.34.ssm_ba.weight q8_0 -blk.34.attn_qkv.weight q8_0 +blk.33.ssm_ba.weight q8_0 +blk.33.ssm_out.weight q8_0 blk.34.attn_gate.weight q8_0 -blk.34.ssm_out.weight q8_0 +blk.34.attn_qkv.weight q8_0 blk.34.ffn_down_shexp.weight q8_0 blk.34.ffn_gate_shexp.weight q8_0 blk.34.ffn_up_shexp.weight q8_0 -blk.35.ffn_down_shexp.weight q8_0 -blk.35.ffn_gate_shexp.weight q8_0 -blk.35.ffn_up_shexp.weight q8_0 +blk.34.ssm_ba.weight q8_0 +blk.34.ssm_out.weight q8_0 blk.35.attn_k.weight q8_0 blk.35.attn_output.weight q8_0 blk.35.attn_q.weight q8_0 blk.35.attn_v.weight q8_0 -blk.36.ssm_ba.weight q8_0 -blk.36.attn_qkv.weight q8_0 +blk.35.ffn_down_shexp.weight q8_0 +blk.35.ffn_gate_shexp.weight q8_0 +blk.35.ffn_up_shexp.weight q8_0 blk.36.attn_gate.weight q8_0 -blk.36.ssm_out.weight q8_0 +blk.36.attn_qkv.weight q8_0 blk.36.ffn_down_shexp.weight q8_0 blk.36.ffn_gate_shexp.weight q8_0 blk.36.ffn_up_shexp.weight q8_0 -blk.37.ssm_ba.weight q8_0 -blk.37.attn_qkv.weight q8_0 +blk.36.ssm_ba.weight q8_0 +blk.36.ssm_out.weight q8_0 blk.37.attn_gate.weight q8_0 -blk.37.ssm_out.weight q8_0 +blk.37.attn_qkv.weight q8_0 blk.37.ffn_down_shexp.weight q8_0 blk.37.ffn_gate_shexp.weight q8_0 blk.37.ffn_up_shexp.weight q8_0 -blk.38.ssm_ba.weight q8_0 -blk.38.attn_qkv.weight q8_0 +blk.37.ssm_ba.weight q8_0 +blk.37.ssm_out.weight q8_0 blk.38.attn_gate.weight q8_0 -blk.38.ssm_out.weight q8_0 +blk.38.attn_qkv.weight q8_0 blk.38.ffn_down_shexp.weight q8_0 blk.38.ffn_gate_shexp.weight q8_0 blk.38.ffn_up_shexp.weight q8_0 -blk.39.ffn_down_shexp.weight q8_0 -blk.39.ffn_gate_shexp.weight q8_0 -blk.39.ffn_up_shexp.weight q8_0 +blk.38.ssm_ba.weight q8_0 +blk.38.ssm_out.weight q8_0 blk.39.attn_k.weight q8_0 blk.39.attn_output.weight q8_0 blk.39.attn_q.weight q8_0 blk.39.attn_v.weight q8_0 -blk.40.ssm_ba.weight q8_0 -blk.40.attn_qkv.weight q8_0 +blk.39.ffn_down_shexp.weight q8_0 +blk.39.ffn_gate_shexp.weight q8_0 +blk.39.ffn_up_shexp.weight q8_0 blk.40.attn_gate.weight q8_0 -blk.40.ssm_out.weight q8_0 +blk.40.attn_qkv.weight q8_0 blk.40.ffn_down_shexp.weight q8_0 blk.40.ffn_gate_shexp.weight q8_0 blk.40.ffn_up_shexp.weight q8_0 -blk.41.ssm_ba.weight q8_0 -blk.41.attn_qkv.weight q8_0 +blk.40.ssm_ba.weight q8_0 +blk.40.ssm_out.weight q8_0 blk.41.attn_gate.weight q8_0 -blk.41.ssm_out.weight q8_0 +blk.41.attn_qkv.weight q8_0 blk.41.ffn_down_shexp.weight q8_0 blk.41.ffn_gate_shexp.weight q8_0 blk.41.ffn_up_shexp.weight q8_0 -blk.42.ssm_ba.weight q8_0 -blk.42.attn_qkv.weight q8_0 +blk.41.ssm_ba.weight q8_0 +blk.41.ssm_out.weight q8_0 blk.42.attn_gate.weight q8_0 -blk.42.ssm_out.weight q8_0 +blk.42.attn_qkv.weight q8_0 blk.42.ffn_down_shexp.weight q8_0 blk.42.ffn_gate_shexp.weight q8_0 blk.42.ffn_up_shexp.weight q8_0 -blk.43.ffn_down_shexp.weight q8_0 -blk.43.ffn_gate_shexp.weight q8_0 -blk.43.ffn_up_shexp.weight q8_0 +blk.42.ssm_ba.weight q8_0 +blk.42.ssm_out.weight q8_0 blk.43.attn_k.weight q8_0 blk.43.attn_output.weight q8_0 blk.43.attn_q.weight q8_0 blk.43.attn_v.weight q8_0 -blk.44.ssm_ba.weight q8_0 -blk.44.attn_qkv.weight q8_0 +blk.43.ffn_down_shexp.weight q8_0 +blk.43.ffn_gate_shexp.weight q8_0 +blk.43.ffn_up_shexp.weight q8_0 blk.44.attn_gate.weight q8_0 -blk.44.ssm_out.weight q8_0 +blk.44.attn_qkv.weight q8_0 blk.44.ffn_down_shexp.weight q8_0 blk.44.ffn_gate_shexp.weight q8_0 blk.44.ffn_up_shexp.weight q8_0 -blk.45.ssm_ba.weight q8_0 -blk.45.attn_qkv.weight q8_0 +blk.44.ssm_ba.weight q8_0 +blk.44.ssm_out.weight q8_0 blk.45.attn_gate.weight q8_0 -blk.45.ssm_out.weight q8_0 +blk.45.attn_qkv.weight q8_0 blk.45.ffn_down_shexp.weight q8_0 blk.45.ffn_gate_shexp.weight q8_0 blk.45.ffn_up_shexp.weight q8_0 -blk.46.ssm_ba.weight q8_0 -blk.46.attn_qkv.weight q8_0 +blk.45.ssm_ba.weight q8_0 +blk.45.ssm_out.weight q8_0 blk.46.attn_gate.weight q8_0 -blk.46.ssm_out.weight q8_0 +blk.46.attn_qkv.weight q8_0 blk.46.ffn_down_shexp.weight q8_0 blk.46.ffn_gate_shexp.weight q8_0 blk.46.ffn_up_shexp.weight q8_0 -blk.47.ffn_down_shexp.weight q8_0 -blk.47.ffn_gate_shexp.weight q8_0 -blk.47.ffn_up_shexp.weight q8_0 +blk.46.ssm_ba.weight q8_0 +blk.46.ssm_out.weight q8_0 blk.47.attn_k.weight q8_0 blk.47.attn_output.weight q8_0 blk.47.attn_q.weight q8_0 blk.47.attn_v.weight q8_0 -output.weight q8_0 +blk.47.ffn_down_shexp.weight q8_0 +blk.47.ffn_gate_shexp.weight q8_0 +blk.47.ffn_up_shexp.weight q8_0 diff --git a/tests/test-quant-type-selection.cpp b/tests/test-quant-type-selection.cpp index ccecbed5c6..6e0f9461c6 100644 --- a/tests/test-quant-type-selection.cpp +++ b/tests/test-quant-type-selection.cpp @@ -5,6 +5,7 @@ #include "ggml-cpp.h" #include "gguf-model-data.h" +#include #include #include #include @@ -269,6 +270,17 @@ static mock_tensors build_mock_tensors(const quantize_state_impl * qs, } } + // sort by layer index then name, matching llama_model_loader::weight_name_comparer + std::sort(result.begin(), result.end(), [](const ggml_tensor * a, const ggml_tensor * b) { + int a_layer = -1, b_layer = -1; + sscanf(a->name, "blk.%d.", &a_layer); + sscanf(b->name, "blk.%d.", &b_layer); + if (a_layer != b_layer) { + return a_layer < b_layer; + } + return strcmp(a->name, b->name) < 0; + }); + return { std::move(ctx), std::move(result) }; }