huggingface/lighteval

Pull Requests Commits

Improve test (#674)

qubvel committed 246 days ago

Verified 2ef1bc53

Update README.md

NathanHB committed 249 days ago

Verified 4ab19e80

Nathan unify modelargs (#609)

NathanHB committed 249 days ago

Verified feb8b706

Parse string values for add_special_tokens in vLLM (#598)

eldarkurtic committed 254 days ago

Verified cc95ff27

update LCB prompt, fix newlines (#645)

rawsh committed 254 days ago

Verified d5efccf3

Lighteval math (#630)

HERIUN committed 254 days ago

Verified 2c3fdf36

Parse seed for vLLM (#602)

eldarkurtic committed 254 days ago

Verified 9ea7d6af

fix slow tests by removing HF_TOKEN (#667)

NathanHB committed 254 days ago

Verified 582d927f

Align AIME pass@1 with literature (#666)

lewtun committed 254 days ago

Verified bb14995c

Fix passing of generation config to main_accelerate (#659)

algo-home committed 255 days ago

Verified 632b69b8

add cot_prompt in vllm (#654)

HERIUN committed 255 days ago

Verified f8ebe644

add uzbek literals (#664)

murodbecks committed 255 days ago

Verified dceb39d6

Better CI and tests (#655)

NathanHB committed 255 days ago

Verified 840a6a31

add arc agi 2 (#642)

NathanHB committed 255 days ago

Verified 75f44298

Use `n=16` samples to estimate `pass@1` for AIME benchmarks (#661)

lewtun committed 255 days ago

Verified f3639c67

Pass At K Math (#647)

clefourrier committed 259 days ago

Verified fcb784d5

Fix docs (#614)

eldarkurtic committed 260 days ago

Verified bfb1099c

Use openai == 1.55.2 and httpx == 0.27.2 (#648)

alozowski committed 262 days ago

Verified 05c5b0ed

Integrate huggingface_hub inference support (#651)

alozowski committed 262 days ago

Verified e611a91f

fix: timedelta import is missing from the lighteval python-api example (#613)

Laz4rz committed 266 days ago

Verified a619b5fa

Rename __initi__.py to __init__.py (#641)

clefourrier committed 266 days ago

Verified b455cd25

Use `blfoat16` as default for vllm models. (#638)

NathanHB committed 266 days ago

Verified 470fa921

Fix steps numbers in custom_yourbench_task.py (#643)

alozowski committed 267 days ago

Verified 2583a0a1

Propagate vLLM batch size controls (#588)

alvin319 committed 268 days ago

Verified bbbdd225

Update README.md (#639)

NathanHB committed 268 days ago

Verified bd9019d6

fix:change multichoice_continuations_start_space category (#610)

AnikiFan committed 269 days ago

Verified d9d90862

enable together models and reasoning models as judges. (#537)

JoelNiklaus committed 269 days ago

Verified d1fcd224

Fixing some silent bugs in Arabic Custom Tasks (#556)

alielfilali01 committed 269 days ago

Verified 706f97eb

support more prequantized models (#620)

DerekLiu35 committed 269 days ago

Verified 35d73d5b

v0.8.1.dev (#637)

NathanHB committed 269 days ago

Verified ecb72067

Newer Older