NielsRogge/transformers

Pull Requests Commits

NielsRogge committed 2 days ago

7ea8148a

Docs: fix typos in Get started (index, quicktour) (#43666)

CodeByKodi committed 4 days ago

Verified 78e4f885

the cache class is deprecated

vasqu committed 4 days ago

a9cf5e53

custom tok init fix (#43591)

itazap committed 4 days ago

Verified f12b50a0

More export friendly rewrites and skipping the failing ones (#43436)

IlyasMoutawwakil committed 4 days ago

Verified 41771ce4

Cast byte_count to int in caching_allocator_warmup for MPS compatibility (#43608)

tobyliu2004 committed 4 days ago

Verified 559af445

[Docs] Complete missing Llama4 configuration docs (#43460)

udaymehta committed 4 days ago

Verified 02943590

Fix t5 failures (#43374)

Abdennacer-Badaoui committed 4 days ago

Verified 07aed9d7

Add EoMT with DINOv3 backbone (#41212)

NielsRogge committed 4 days ago

Verified 751cff7c

Update DBRX docs to reference re-uploaded checkpoint (#43196)

qgallouedec committed 4 days ago

Verified 645db400

[loading] Fix forced upcasting to fp32 (#43683)

Cyrilvallez committed 4 days ago

Verified fb22e73f

Fix FP8Expert for Qwen (#43670)

yiliu30 committed 4 days ago

Verified 010b9f1e

Simplify loading structure (#43589)

Cyrilvallez committed 4 days ago

Verified 0fa2c2f0

[CB] Refactor logic for inputs and outputs outside of the main API (#43569)

remi-or committed 4 days ago

Verified e5c8b0d1

Make sure hub errors are surfaced in `PreTrainedTokenizerBase` (#43675)

tarekziade committed 4 days ago

Verified 5ccea230

Fix `FP8Expert` for DeepSeek R1 (#43616)

yiliu30 committed 4 days ago

Verified 8fce0a18

Use correct sampling rate in chat template (#43674)

zucchini-nlp committed 4 days ago

Verified e4b22baa

[`HunYuan`] Fix RoPE init (#43411)

vasqu committed 4 days ago

Verified 4cd791d6

XPU now supports MoE kernel(MegaBlocks) implementation (#43435)

YangKai0616 committed 4 days ago

Verified b9f0074f

[`Sam`] Fixup training flags (#43567)

vasqu committed 4 days ago

Verified aeb728ab

remove torchao.autoquant from transformers (#43561)

vkuzo committed 4 days ago

Verified 5772c733

[DeepSpeed] properly handle MoE weight conversion (#43524)

kashif committed 4 days ago

Verified 063b4569

Tie zamba weights correctly (#43623)

zucchini-nlp committed 4 days ago

Verified d5eab539

[kernels] Centralize kernels tests (#42819)

MekkCyber committed 4 days ago

Verified 4da99340

Fix `process_bad_commit_report.py`: avoid items to appear in `null` author in the report (#43662)

ydshieh committed 6 days ago

Verified 78bb8514

Fix `KeyError` in `check_bad_commit.py` (#43655)

ydshieh committed 6 days ago

Verified 8c403da7

[Benchmark] Minor fix for benchmark: kernel is not correctly called (#43428)

sywangyi committed 6 days ago

Verified 17221633

Add explicit commit info to PR comment CI feedback (#43635)

ydshieh committed 6 days ago

Verified 2ff1d4f0

Better new failures reporting for PR comment CI (#43629)

ydshieh committed 7 days ago

Verified 5a67f0a7

[docs] serving (#42853)

stevhliu committed 7 days ago

Verified 94848a13

Older