Default Branch

main 8447d70b18 · Some gemma 3 improvements (#1000) · Updated 2026-04-06 10:05:05 +08:00

ZIP  TAR.GZ

Branches

debug-exclude-newer fd11713ed9 · exclude torchvission from nightly · Updated 2026-04-05 01:17:50 +08:00    git-market	2 2		ZIP  TAR.GZ
dockerfile 9f7dbb2493 · Update docker file · Updated 2025-10-07 07:31:59 +08:00    git-market	76 1		ZIP  TAR.GZ
requirements-update b1f852c1ba · Update requirements.txt · Updated 2025-09-27 10:57:22 +08:00    git-market	84 2		ZIP  TAR.GZ
qwen-tokenizer-fix 862df48e38 · use apply_chat_template · Updated 2025-09-16 21:12:01 +08:00    git-market	91 9		ZIP  TAR.GZ
gemma-3 8fd29ed079 · Gemma 3 270M from scratch · Updated 2025-08-17 08:49:38 +08:00    git-market	804 679		ZIP  TAR.GZ
fix-qwen3-tokeniizer-eos 06aa6d470a · Fix eos token usage in Qwen3 tokenizer · Updated 2025-08-06 02:42:18 +08:00    git-market	128 1		ZIP  TAR.GZ
head-dim 4aa398c79d · Comment typo: head_dim -> head_dim // 2 · Updated 2025-07-23 21:16:30 +08:00    git-market	131 1		ZIP  TAR.GZ
weezymatt-bugfix/semantic-error 1552023bd4 · Fix issue 724: unused args (#726) · Updated 2025-07-11 01:58:32 +08:00    git-market	143 9		ZIP  TAR.GZ
fix_masking_kv_cache 713a6e24c9 · add tests · Updated 2025-06-23 06:48:23 +08:00    git-market	162 2		ZIP  TAR.GZ
cleaning 4715dc3be5 · remove redundant context_length in GQA · Updated 2025-04-01 05:49:10 +08:00    git-market	206 3		ZIP  TAR.GZ
uv-improvements ca0eee4cf9 · simplify and use pythorch 3.12 · Updated 2025-02-20 11:01:15 +08:00    git-market	238 7		ZIP  TAR.GZ