huggingface/transformers

Pull Requests Commits

Fix Gemma3 embedding scaling (#37109)

gau-nernst committed 1 year ago

Verified 19919689

[MLU] Fix FA2 check error, remove deepspeed-mlu deps. (#36159)

huismiling committed 1 year ago

Verified d0b65bb4

fix whisper re-compile (#36712)

jiqing-feng committed 1 year ago

Verified ad63d20d

enable tp on CPU (#36299)

jiqing-feng committed 1 year ago

Verified 286393fb

Fix 4090/ada not detected as having FP8 support (#37067)

Qubitium committed 1 year ago

Verified 4705b04c

Support passing flash_attn_kwargs when gradient_checkpointing is enabled (#37037)

efsotr committed 1 year ago

Verified 2b4734bd

Gaudi: Fix the pipeline failed issue with hpu device (#36990)

yuanwu2017 committed 1 year ago

Verified bd41b9c1

Adding Qwen3 and Qwen3MoE (#36878)

bozheng-hit committed 1 year ago

Verified 6acd5aec

🌐 [i18n-KO] Translated `qwen2_vl.md` to Korean (#36750)

MinJu-Ha committed 1 year ago

Verified 0d6a60fe

ydshieh committed 1 year ago

Verified b7fc2daf

[Cache] rename dtype attribute 🚨 🚨 (#37044)

gante committed 1 year ago

Verified bab605dd

[generate] beam search -- fix output cropping (#37080)

gante committed 1 year ago

Verified 9fd94760

fixed typo. (#37057)

湛露先生 committed 1 year ago

Verified 257bc670

Fix AttentionInterface following feedback (#37010)

Cyrilvallez committed 1 year ago

Verified 2bea6bf2

Fix state_dict map location when quantized (#37086)

Cyrilvallez committed 1 year ago

Verified a86dad56

Update w/ new account (#37084)

muellerzr committed 1 year ago

Verified d6064754

fix tied weigths issue (#37031)

ydshieh committed 1 year ago

Verified 581cf96e

[WIP] add deepseek-v3 (#35926)

bzantium committed 1 year ago

Verified eca74d13

[blip-2] Fix dtype mismatch when keep in fp32 (#37068)

zucchini-nlp committed 1 year ago

Verified 52cc204d

Change deprecated PT functions (#37041)

cyyever committed 1 year ago

Verified aa3778af

Fix some typos about benchmark scripts. (#37027)

湛露先生 committed 1 year ago

Verified c90e6e96

Use `lru_cache` for tokenization tests (#36818)

ydshieh committed 1 year ago

Verified 1fcaad6d

fix: AttributeError: 'LlavaProcessor' object has no attribute 'image_token_id' (#37026)

jp1924 committed 1 year ago

Verified 3af425d4

Fix SDPA implementation in Qwen2-VL (issues with torch==2.6.0) (#36891)

ManuelFay committed 1 year ago

Verified 064cd7cd

fix: Fully remove legacy cache from Llama (#36958)

Wheest committed 1 year ago

Verified 348f3285

fixed typo (#37036)

finnoh committed 1 year ago

Verified d6b3c748

Remove deprecated batch_size parameter (#37007)

cyyever committed 1 year ago

Verified 6cc9c8d7

Replace default split function with jnp.split() in flax models (#37001)

premmurugan229 committed 1 year ago

Verified 4cc65e99

Set weights_only in torch.load (#36991)

cyyever committed 1 year ago

Verified 41a0e58e

Fix typing for None valued variables (#37004)

cyyever committed 1 year ago

Verified de77f5b1

Newer Older