license: apache-2.0 language: am
DeepSpeed-RLHF系统训练：DeepSpeed-HE 能够在 RLHF 中无缝地在推理和训练模式之间切换，使其能够利用来自 DeepSpeed-Inference 的各种优化，如张量并行计算和高性能CUDA算子进行语言生成，同时对训练部分还能从 ZeRO- 和 LoRA-based 内存优化策略中受益。DeepSpeed-HE 还能够自动在 RLHF 的不同阶段进行智能的内存管理和数据缓存。
Train Data：（English）--data_path Dahoas/rm-static Dahoas/full-hh-rlhf Dahoas/synthetic-instruct-gptj-pairwise yitingxie/rlhf-reward-datasets openai/webgpt_comparisons stanfordnlp/SHP
Train Data：（Chinese）--data_path wangrui6/Zhihu-KOL Cohere/miracl-zh-queries-22-12 Hello-SimpleAI/HC3-Chinese mkqa-Chinese
可自定义actor model 和 reward model，亦可单独训练rlhf model
Usage:
git clone https://github.com/microsoft/DeepSpeedExamples cd DeepSpeedExamples/applications/DeepSpeed-Chat pip install -r requirements.txt python chat.py --path Laurie/opt1.3b-deepspeed-chat
- Downloads last month
- 23
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.