by stanleylsx · AI Tool · ★ 223
Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h
LLMs Tool Introduction 一个基于🤗HuggingFace开发的大语言模型训练、测试工具。支持不同模型的webui、终端预测，支持各模型的低参数量及全参数模型的预训练、奖励模型训练以及RLHF训练(PPO和DPO两种方法)。同时支持deepspeed分布式训练。 作者习惯于把配置和要做的事情都写在一个配置文件里面，然后以一个主函数作为入口直接运行，所以才有了这个项目，喜欢指令的朋友们可以改回去使用。 Updates Date| Detail :| 2023-10-30|通过attentionsinks支持StreamingLLM 2023-10-25|基于sentencepiece实现词表扩充功能 2023-10-24|支持使用NEFTune对LLM进行noise tune 2023-10-09|增加扩充词表后Embedding初始化方式 2023-10-08|LLama和Falcon两类模型支持Flash Attention2 2023-09-26|支持模型预训练 2023-09-11|多轮对话的Firefly的loss训练函数集成 2023-09-04|支持部分可以从配置修改使用NTK的模型 2023-08-24|支持deepspeed-ZeRo2分布式训练 2023-08-23|RLHF的DPO方法对各个模型的训练支持 2023-08-21|RLHF的PPO方法对各个模型的训练支持 2023-08-08|奖励模型训练 2023-07-25|初始仓库 Requirement 几个重要环境： python：3.10+ torch：2.0.1+ bitsandbytes：不同操作系统下需要对应安装不同的包（Linux下0.39.0+，Windows下要专门下载对应的wheel本地安装） 其它环境见requirements.txt 目前FlashAttention作者未主动兼容和测试Windows操作环境issues，若在Windows上不用安装flash-attn这个包。 Feature Supported models 大模型经过SFT(然后做RLHF)之后可用于对话任务C
| Stars | 223 |
| Forks | 21 |
| Language | Python |
| Category | AI Tool |
| License | Apache-2.0 |
| Quality Score | 34.2/100 |
| Open Issues | 2 |
| Last Updated | 2023-12-08 |
| Created | 2023-07-20 |
| Platforms | browser, python |
| Est. Tokens | ~657k |
Looking for a llms_tool alternative? If you're comparing llms_tool with other ai tool tools, these 6 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.
Awesome-LLM-Eval: a curated list of tools, datasets/benchmark, demos, leaderboard, papers, docs and models, ma
Toolkit for fine-tuning, ablating and unit-testing open-source LLMs.
A Reactive CLI that generates commit messages for Git and Jujutsu with Ollama, ChatGPT, Gemini, Claude, Mistra
WorkflowAI is an open-source platform where product and engineering teams collaborate to build and iterate on
Large Language Models (LLMs) applications and tools running on Apple Silicon in real-time with Apple MLX.
The open creative AI workspace
Explore other popular ai tool tools:
llms_tool is 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。. It is categorized as a AI Tool with 223 GitHub stars.
llms_tool is primarily written in Python. It covers topics such as aquila, aquila2, baichuan.
You can find installation instructions and usage details in the llms_tool GitHub repository at github.com/stanleylsx/llms_tool. The project has 223 stars and 21 forks, indicating an active community.
llms_tool is released under the Apache-2.0 license, making it free to use and modify according to the license terms.
The top alternatives to llms_tool on Agent Skills Hub include Awesome-LLM-Eval, LLM-Finetuning-Toolkit, aicommit2. Each offers a different approach to the same problem space — compare them side-by-side by stars, quality score, and community activity.