HappyZ happyz
happyz synced commits to refs/pull/7458/merge at happyz/llama.cpp from mirror 2024-05-25 11:05:57 -07:00
3cbd23ed88 labeler: added Apple Metal detector (+Kompute) (#7529)
00c6390793 main : don't print special tokens with --grammar (#6923)
faa0e6979a ggml: aarch64: SVE kernels for q8_0_q8_0, q4_0_q8_0 vector dot (#7433)
9791f40258 android : module (#7502)
Compare 32 commits »
happyz synced commits to refs/pull/7436/merge at happyz/llama.cpp from mirror 2024-05-25 11:05:56 -07:00
8bef9c5b66 Merge 32ba8d3e20b392eb638e57adaa974357dda7b392 into 9588f196b1
9588f196b1 train : change default FA argument (#7528)
3cbd23ed88 labeler: added Apple Metal detector (+Kompute) (#7529)
00c6390793 main : don't print special tokens with --grammar (#6923)
faa0e6979a ggml: aarch64: SVE kernels for q8_0_q8_0, q4_0_q8_0 vector dot (#7433)
Compare 9 commits »
happyz synced commits to refs/pull/7455/merge at happyz/llama.cpp from mirror 2024-05-25 11:05:56 -07:00
9588f196b1 train : change default FA argument (#7528)
3cbd23ed88 labeler: added Apple Metal detector (+Kompute) (#7529)
00c6390793 main : don't print special tokens with --grammar (#6923)
faa0e6979a ggml: aarch64: SVE kernels for q8_0_q8_0, q4_0_q8_0 vector dot (#7433)
Compare 9 commits »
happyz synced commits to refs/pull/7414/merge at happyz/llama.cpp from mirror 2024-05-25 11:05:56 -07:00
9588f196b1 train : change default FA argument (#7528)
3cbd23ed88 labeler: added Apple Metal detector (+Kompute) (#7529)
00c6390793 main : don't print special tokens with --grammar (#6923)
faa0e6979a ggml: aarch64: SVE kernels for q8_0_q8_0, q4_0_q8_0 vector dot (#7433)
Compare 7 commits »
happyz synced commits to refs/pull/7402/merge at happyz/llama.cpp from mirror 2024-05-25 11:05:56 -07:00
9588f196b1 train : change default FA argument (#7528)
3cbd23ed88 labeler: added Apple Metal detector (+Kompute) (#7529)
00c6390793 main : don't print special tokens with --grammar (#6923)
faa0e6979a ggml: aarch64: SVE kernels for q8_0_q8_0, q4_0_q8_0 vector dot (#7433)
Compare 6 commits »
happyz synced commits to refs/pull/7430/merge at happyz/llama.cpp from mirror 2024-05-25 11:05:56 -07:00
9588f196b1 train : change default FA argument (#7528)
3cbd23ed88 labeler: added Apple Metal detector (+Kompute) (#7529)
00c6390793 main : don't print special tokens with --grammar (#6923)
faa0e6979a ggml: aarch64: SVE kernels for q8_0_q8_0, q4_0_q8_0 vector dot (#7433)
Compare 7 commits »
happyz synced commits to refs/pull/7435/merge at happyz/llama.cpp from mirror 2024-05-25 11:05:56 -07:00
08ae318a81 Merge 38280a6afe4407b2fd727b5de66da416a6c18b52 into 9588f196b1
9588f196b1 train : change default FA argument (#7528)
3cbd23ed88 labeler: added Apple Metal detector (+Kompute) (#7529)
00c6390793 main : don't print special tokens with --grammar (#6923)
faa0e6979a ggml: aarch64: SVE kernels for q8_0_q8_0, q4_0_q8_0 vector dot (#7433)
Compare 9 commits »
happyz synced commits to refs/pull/7246/merge at happyz/llama.cpp from mirror 2024-05-25 11:05:55 -07:00
902184dd3a fix missing slash in `fs_get_cache_directory()` (#7503)
57684331fc Make tokenize CLI tool have nicer command line arguments. (#6188)
b83bab15a5 gguf-py : fix and simplify quantized shape round-trip (#7483)
d041d2ceaa flake.lock: Update (#7232)
Compare 7 commits »
happyz synced commits to refs/pull/7267/merge at happyz/llama.cpp from mirror 2024-05-25 11:05:55 -07:00
9588f196b1 train : change default FA argument (#7528)
3cbd23ed88 labeler: added Apple Metal detector (+Kompute) (#7529)
00c6390793 main : don't print special tokens with --grammar (#6923)
faa0e6979a ggml: aarch64: SVE kernels for q8_0_q8_0, q4_0_q8_0 vector dot (#7433)
Compare 34 commits »
happyz synced commits to refs/pull/7239/merge at happyz/llama.cpp from mirror 2024-05-25 11:05:55 -07:00
9588f196b1 train : change default FA argument (#7528)
3cbd23ed88 labeler: added Apple Metal detector (+Kompute) (#7529)
00c6390793 main : don't print special tokens with --grammar (#6923)
faa0e6979a ggml: aarch64: SVE kernels for q8_0_q8_0, q4_0_q8_0 vector dot (#7433)
Compare 13 commits »
happyz synced commits to refs/pull/7187/merge at happyz/llama.cpp from mirror 2024-05-25 11:05:55 -07:00
2296837e3d Merge c9b2297de0a23bbb2af99cb046cf38682438ba74 into 902184dd3a
902184dd3a fix missing slash in `fs_get_cache_directory()` (#7503)
57684331fc Make tokenize CLI tool have nicer command line arguments. (#6188)
b83bab15a5 gguf-py : fix and simplify quantized shape round-trip (#7483)
Compare 4 commits »
happyz synced commits to refs/pull/7342/merge at happyz/llama.cpp from mirror 2024-05-25 11:05:55 -07:00
9588f196b1 train : change default FA argument (#7528)
3cbd23ed88 labeler: added Apple Metal detector (+Kompute) (#7529)
00c6390793 main : don't print special tokens with --grammar (#6923)
faa0e6979a ggml: aarch64: SVE kernels for q8_0_q8_0, q4_0_q8_0 vector dot (#7433)
Compare 6 commits »
happyz synced commits to refs/pull/7300/merge at happyz/llama.cpp from mirror 2024-05-25 11:05:55 -07:00
9588f196b1 train : change default FA argument (#7528)
3cbd23ed88 labeler: added Apple Metal detector (+Kompute) (#7529)
00c6390793 main : don't print special tokens with --grammar (#6923)
faa0e6979a ggml: aarch64: SVE kernels for q8_0_q8_0, q4_0_q8_0 vector dot (#7433)
Compare 11 commits »
happyz synced commits to refs/pull/7058/merge at happyz/llama.cpp from mirror 2024-05-25 11:05:55 -07:00
9588f196b1 train : change default FA argument (#7528)
739648f3e6 Implement Q8_0 quantization fully in PyTorch.
abc958b07e Move noqa comment to where the lastest flake8 likes it.
0a1ef1127f Write tensors in layer order.
Compare 22 commits »
happyz synced commits to refs/pull/6784/merge at happyz/llama.cpp from mirror 2024-05-25 11:05:54 -07:00
9588f196b1 train : change default FA argument (#7528)
3cbd23ed88 labeler: added Apple Metal detector (+Kompute) (#7529)
00c6390793 main : don't print special tokens with --grammar (#6923)
faa0e6979a ggml: aarch64: SVE kernels for q8_0_q8_0, q4_0_q8_0 vector dot (#7433)
Compare 9 commits »
happyz synced commits to refs/pull/6839/merge at happyz/llama.cpp from mirror 2024-05-25 11:05:54 -07:00
9588f196b1 train : change default FA argument (#7528)
3cbd23ed88 labeler: added Apple Metal detector (+Kompute) (#7529)
00c6390793 main : don't print special tokens with --grammar (#6923)
faa0e6979a ggml: aarch64: SVE kernels for q8_0_q8_0, q4_0_q8_0 vector dot (#7433)
Compare 6 commits »
happyz synced commits to refs/pull/6840/merge at happyz/llama.cpp from mirror 2024-05-25 11:05:54 -07:00
099f34d45d Merge 0629a79d032d1dbfbc25a6ef40e830e895ed2660 into 9588f196b1
9588f196b1 train : change default FA argument (#7528)
3cbd23ed88 labeler: added Apple Metal detector (+Kompute) (#7529)
00c6390793 main : don't print special tokens with --grammar (#6923)
faa0e6979a ggml: aarch64: SVE kernels for q8_0_q8_0, q4_0_q8_0 vector dot (#7433)
Compare 8 commits »
happyz synced commits to refs/pull/6844/merge at happyz/llama.cpp from mirror 2024-05-25 11:05:54 -07:00
9588f196b1 train : change default FA argument (#7528)
3cbd23ed88 labeler: added Apple Metal detector (+Kompute) (#7529)
00c6390793 main : don't print special tokens with --grammar (#6923)
faa0e6979a ggml: aarch64: SVE kernels for q8_0_q8_0, q4_0_q8_0 vector dot (#7433)
Compare 9 commits »
happyz synced commits to refs/pull/6923/head at happyz/llama.cpp from mirror 2024-05-25 11:05:54 -07:00
e75c5ca451 main: remove special token file descriptor feature (#5)
happyz synced commits to refs/pull/6940/merge at happyz/llama.cpp from mirror 2024-05-25 11:05:54 -07:00
9588f196b1 train : change default FA argument (#7528)
3cbd23ed88 labeler: added Apple Metal detector (+Kompute) (#7529)
00c6390793 main : don't print special tokens with --grammar (#6923)
faa0e6979a ggml: aarch64: SVE kernels for q8_0_q8_0, q4_0_q8_0 vector dot (#7433)
Compare 9 commits »