Miraclemarvel55 / ChatGLM-RLHFLinks
对ChatGLM直接使用RLHF提升或降低目标输出概率|Modify ChatGLM output with only RLHF
☆195Updated 2 years ago
Alternatives and similar repositories for ChatGLM-RLHF
Users that are interested in ChatGLM-RLHF are comparing it to the libraries listed below
Sorting:
- Implementation of Chinese ChatGPT☆288Updated last year
- 用于大模型 RLHF 进行人工数据标注排序的工具。A tool for manual response data annotation sorting in RLHF stage.☆254Updated 2 years ago
- alpaca中文指令微调数据集☆395Updated 2 years ago
- ☆308Updated 2 years ago
- chatglm多gpu用deepspeed和☆412Updated last year
- deep learning☆149Updated 5 months ago
- ChatGLM2-6B 全参数微调,支持多轮对话的高效微调。☆401Updated 2 years ago
- Firefly中文LLaMA-2大模型,支持 增量预训练Baichuan2、Llama2、Llama、Falcon、Qwen、Baichuan、InternLM、Bloom等大模型☆413Updated last year
- ☆281Updated last year
- Baichuan-13B 指令微调☆90Updated 2 years ago
- chatglm-6b微调/LORA/PPO/推理, 样本为自动生成的整数/小数加减乘除运算, 可gpu/cpu☆164Updated 2 years ago
- 一个基于HuggingFace开发的大语言模型训练、测试工具。支持各模型的webui、终端预测,低参数量及全参数模型训练(预训练、SFT、RM、PPO、DPO)和融合、量化。☆220Updated last year
- ChatGLM2-6B微调, SFT/LoRA, instruction finetune☆110Updated 2 years ago
- pCLUE: 1000000+多任务提示学习数据集☆500Updated 3 years ago
- 大语言模型指令调优工具(支持 FlashAttention)☆178Updated last year
- 专注于中文领域大语言模型,落地到某个行业某个领域,成为一个行业大模型、公司级别或行业级别领域大模型。☆123Updated 7 months ago
- A full pipeline to finetune ChatGLM LLM with LoRA and RLHF on consumer hardware. Implementation of RLHF (Reinforcement Learning with Huma…☆138Updated 2 years ago
- deepspeed+trainer简单高效实现多卡微调大模型☆128Updated 2 years ago
- 语言模型中文认知能力分析☆236Updated 2 years ago
- ChatGLM-6B fine-tuning.☆136Updated 2 years ago
- ☆163Updated 2 years ago
- 微调ChatGLM☆128Updated 2 years ago
- 探索中文instruct数据在ChatGLM, LLaMA上的微调表现☆389Updated 2 years ago
- OpenLLMWiki: Docs of OpenLLMAI. Survey, reproduction and domain/task adaptation of open source chatgpt alternatives/implementations. PiXi…☆262Updated 10 months ago
- ChatGLM-6B添加了RLHF的实现,以及部分核心代码的逐行讲解 ,实例部分是做了个新闻短标题的生成,以及指定context推荐的RLHF的实现☆88Updated 2 years ago
- 中文 Instruction tuning datasets☆136Updated last year
- 使用peft库,对chatGLM-6B/chatGLM2-6B实现4bit的QLoRA高效微调,并做lora model和base model的merge及4bit的量化(quantize)。☆357Updated 2 years ago
- ChatGLM-6B 指令学习|指令数据|Instruct☆654Updated 2 years ago
- ☆44Updated last year
- Chinese large language model base generated through incremental pre-training on Chinese datasets☆238Updated 2 years ago