huggingface/transformers

Pull Requests Commits

Merge branch 'main' of github.com:huggingface/transformers into add-deci-lm

ArthurZucker committed 283 days ago

ac0a7157

A bit of cleaning 🧹🧹 (#37215)

Cyrilvallez committed 283 days ago

Verified cdfb018d

Use Python 3.9 syntax in tests (#37343)

cyyever committed 283 days ago

Verified 1e6b546e

convert float for yarn related arguments in rope_scaling (#37139)

bzantium committed 283 days ago

Verified 0fc683d1

Expose blip2qformer (#37254)

alex-jw-brooks committed 283 days ago

Verified 2515a5a2

Multiple llama4 fixe (#37353)

ArthurZucker committed 283 days ago

Verified 2da82e43

Fixing flex attention for torch=2.6.0 (#37285)

salmanmohammadi committed 283 days ago

Verified 794fde7b

more fixes for post-training llama4 (#37329)

winglian committed 283 days ago

Verified b54c2f46

Remove unnecessary attr assignment (#36837)

tugsbayasgalan committed 283 days ago

Verified 754a370b

Updated Model-card for donut (#37290)

Logeswaran7 committed 283 days ago

Verified 31a62c2e

Add bnb to the list of supported quantization methods for LLama4 (#37348)

MekkCyber committed 283 days ago

Verified f8301051

Update Model Card for Jamba (#37152)

ParagEkbote committed 283 days ago

Verified e2b0224d

Improvements in Gemma2 model card (#37076)

devesh-2002 committed 283 days ago

Verified 6cc109c3

Clean up the compressed-tensors integration (#37349)

MekkCyber committed 283 days ago

Verified 8bbcdf54

Update Model card for GPT2 (#37101)

ash-01xor committed 283 days ago

Verified 3a826a45

Update falcon mamba card (#37253)

ricalanis committed 283 days ago

Verified 5e855095

Update model-card for DINOv2 (#37104)

shubham0204 committed 283 days ago

Verified 416b5a87

updated model card for Mistral (#37156)

NahieliV committed 283 days ago

Verified f8a16805

Remove HQQ from caching allocator warmup (#37347)

Cyrilvallez committed 283 days ago

Verified 48e17985

Update translation template (#37294)

stevhliu committed 283 days ago

Verified 832cb684

fix derived berts `_init_weights` (#37341)

Cyrilvallez committed 283 days ago

Verified 22065bd6

Avoid build crashes when torch.version.xpu doesn't exist and fix Llama4 processor tests (#37346)

Rocketknight1 committed 283 days ago

Verified f789f960

enable 2 llama UT cases on xpu (#37126)

yao-matrix committed 284 days ago

Verified 12bf24d6

byebye torch 2.0 (#37277)

ydshieh committed 284 days ago

Verified e7ad0770

Fix torchao usage (#37034)

jiqing-feng committed 284 days ago

Verified 99f9f104

Use Python 3.9 syntax in examples (#37279)

cyyever committed 284 days ago

Verified 0fb8d49e

Fix `init empty weights` without accelerate (#37337)

Cyrilvallez committed 284 days ago

Verified 08f36771

Fix deepspeed with quantization (#37324)

Cyrilvallez committed 284 days ago

Verified 9db31ea5

fix llama4 training (#37319)

hiyouga committed 284 days ago

Verified debfe904

fix flex attn when optional args aren't passed (#37327)

winglian committed 284 days ago

Verified 54538ebe

Older