yanqiangmiffy / how-to-train-tokenizer
怎么训练一个LLM分词器
☆144Updated last year
Alternatives and similar repositories for how-to-train-tokenizer
Users that are interested in how-to-train-tokenizer are comparing it to the libraries listed below
Sorting:
- 使用单个24G显卡，从0开始训练LLM☆53Updated this week
- 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。☆117Updated 2 months ago
- 大语言模型指令调优工具（支持 FlashAttention）☆172Updated last year
- Baichuan2代码的逐行解析版本，适合小白☆213Updated last year
- 中文 Instruction tuning datasets☆131Updated last year
- deepspeed+trainer简单高效实现多卡微调大模型☆125Updated last year
- ☆160Updated 2 years ago
- ChatGLM2-6B微调, SFT/LoRA, instruction finetune☆107Updated last year
- ☆168Updated last year
- ☆84Updated last year
- Finetuning LLaMA with RLHF (Reinforcement Learning with Human Feedback) based on DeepSpeed Chat☆115Updated last year
- Clustering and Ranking: Diversity-preserved Instruction Selection through Expert-aligned Quality Estimation☆77Updated 6 months ago
- The complete training code of the open-source high-performance Llama model, including the full process from pre-training to RLHF.☆65Updated 2 years ago
- ☆97Updated last year
- 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。☆217Updated last year
- baichuan LLM surpervised finetune by lora☆63Updated last year
- ☆63Updated 2 years ago
- 文本去重☆71Updated 11 months ago
- MEASURING MASSIVE MULTITASK CHINESE UNDERSTANDING☆87Updated last year
- 欢迎来到 "LLM-travel" 仓库！探索大语言模型（LLM）的奥秘 🚀。致力于深入理解、探讨以及实现与大模型相关的各种技术、原理和应用。☆321Updated 9 months ago
- ☆226Updated last year
- ☆109Updated 6 months ago
- 使用sentencepiece中BPE训练中文词表，并在transformers中进行使用。☆117Updated last year
- 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☆83Updated last year
- ☆140Updated last year
- ☆67Updated last year
- large language model training-3-stages+deployment☆48Updated last year
- 中文大模型微调(LLM-SFT), 数学指令数据集MWP-Instruct, 支持模型(ChatGLM-6B, LLaMA, Bloom-7B, baichuan-7B), 支持(LoRA, QLoRA, DeepSpeed, UI, TensorboardX), 支持(微…☆199Updated 11 months ago
- 1st Solution For Conversational Multi-Doc QA Workshop & International Challenge @ WSDM'24 - Xiaohongshu.Inc☆160Updated last year
- Baichuan-13B 指令微调☆90Updated last year