builtin_pellm_models.md

June 27, 2024 ยท View on GitHub

Builtin PELLM Models

FATE-LLM provide some builtin pellm models, users can use them simply to efficiently train their language models. To use these models, please read the using tutorial of ChatGLM-6B Training Guide.
After reading the training tutorial above, it's easy to use other models listing in the following tabular by changing module_name, class_name, dataset list below.

ModelModuleNameClassNameDataSetName
Qwen2pellm.qwenQwenprompt_dataset
Bloom-7B1pellm.bloomBloomprompt_dataset
OPT-6.7Bpellm.optOPTprompt_dataset
LLaMA-2-7Bpellm.llamaLLaMaprompt_dataset
LLaMA-7Bpellm.llamaLLaMaprompt_dataset
ChatGLM3-6Bpellm.chatglmChatGLMprompt_dataset
GPT-2pellm.gpt2GPT2CLMprompt_dataset
GPT-2pellm.gpt2GPT2seq_cls_dataset
ALBERTpellm.albertAlbertseq_cls_dataset
BARTpellm.bartBartseq_cls_dataset
BERTpellm.bertBertseq_cls_dataset
DeBERTapellm.debertaDebertaseq_cls_dataset
DistilBERTpellm.distilbertDistilBertseq_cls_dataset
RoBERTapellm.robertaRobertaseq_cls_dataset