数据集:

3Aallenai/c4 3Aallenai/s2orc 3Abigcode/the-stack-dedup 3Acompetition_math 3Aconceptofmind/cot_submix_original/cot_gsm8k 3Aduorc 3Aemozilla/quality 3Aknkarthick/dialogsum 3Amc4 3Amosaicml/dolly_hhrlhf 3Ascrolls/summ_screen_fd 3Aspider 3Atau/scrolls/qasper 3Atogethercomputer/RedPajama-Data-1T

预印本库:

arxiv:1909.08053 arxiv:2010.04245 arxiv:2108.12409 arxiv:2205.14135 arxiv:2302.06675 arxiv:2302.13971

其他:

Composer custom_code llm-foundry MosaicML mpt StreamingDatasets

许可:

apache-2.0 cc-by-sa-3.0

类库:

PyTorch Transformers

任务:

文本生成