diff --git a/tools/mtmd/clip.cpp b/tools/mtmd/clip.cpp index abdb778f7a..52ea542dec 100644 --- a/tools/mtmd/clip.cpp +++ b/tools/mtmd/clip.cpp @@ -1175,10 +1175,11 @@ struct clip_graph { cb(K, "resampler_K", -1); cb(V, "resampler_V", -1); + float resampler_kq_scale = 1.0f/ sqrtf(float(d_head)); embeddings = build_attn( model.mm_model_attn_o_w, model.mm_model_attn_o_b, - Q, K, V, nullptr, kq_scale, -1); + Q, K, V, nullptr, resampler_kq_scale, -1); cb(embeddings, "resampler_attn_out", -1); } // layernorm