CrazyBoyM / LLM-Chinese
(撰写ing..)本仓库偏教程性质,以「模型中文化」为一个典型的模型训练问题切入场景,指导读者上手学习LLM二次微调训练。
☆34Updated 9 months ago
Alternatives and similar repositories for LLM-Chinese:
Users that are interested in LLM-Chinese are comparing it to the libraries listed below
- This repository provides an implementation of the paper "A Simple yet Effective Training-free Prompt-free Approach to Chinese Spelling Co…☆68Updated 2 months ago
- A demo built on Megrez-3B-Instruct, integrating a web search tool to enhance the model's question-and-answer capabilities.☆37Updated 4 months ago
- 大语言模型训练和服务调研☆37Updated last year
- the newest version of llama3,source code explained line by line using Chinese☆22Updated last year
- 文本去重☆71Updated 11 months ago
- 国内首个全参数训练的法律大模型 HanFei-1.0 (韩非)☆116Updated last year
- 中文原生检索增强生成测评基准☆115Updated last year
- XVERSE-65B: A multilingual large language model developed by XVERSE Technology Inc.☆140Updated last year
- 中文基于满血DeepSeek-R1蒸馏数据集☆54Updated 2 months ago
- (1)弹性区间标准化的旋转位置词嵌入编码器+peft LORA量化训练,提高万级tokens性能支持。(2)证据理论解释学习,提升模型的复杂逻辑推理能力(3)兼容alpaca数据格式。☆44Updated last year
- ChatGLM2-6B微调, SFT/LoRA, instruction finetune☆107Updated last year
- 使用langchain进行任务规划,构建子任务的会话场景资源,通过MCTS任务执行器,来让每个子任务通过在上下文中资源,通过自身反思探索来获取自身对问题的最优答案;这种方式依赖模型的对齐偏好,我们在每种偏好上设计了一个工程框架,来完成自我对不同答案的奖励进行采样策略☆29Updated 3 weeks ago
- SUS-Chat: Instruction tuning done right☆48Updated last year
- 我们是第一个完全可商用的角色大模型。☆40Updated 8 months ago
- 基于baichuan-7b的开源多模态大语言模型☆73Updated last year
- 首个llama2 13b 中文版模型 (Base + 中文对话SFT,实现流畅多轮人机自然语言交互)☆90Updated last year
- Light local website for displaying performances from different chat models.☆86Updated last year
- ☆65Updated 7 months ago
- SearchGPT: Building a quick conversation-based search engine with LLMs.☆46Updated 4 months ago
- Imitate OpenAI with Local Models☆88Updated 8 months ago
- Evaluation for AI apps and agent☆41Updated last year
- 本项目旨在对大量文本文件进行快速编码检测和转换以辅助mnbvc语料集项目的数据清洗工作☆61Updated 6 months ago
- ☆46Updated 10 months ago
- 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。☆118Updated 2 months ago
- Qwen-WisdomVast is a large model trained on 1 million high-quality Chinese multi-turn SFT data, 200,000 English multi-turn SFT data, and …☆18Updated last year
- qwen models finetuning☆97Updated 2 months ago
- ☆40Updated last year
- ☆227Updated 11 months ago
- Reproduction paper --- PDFTriage : Question Answering over Long, Structured Documents☆40Updated last year
- Alpaca Chinese Dataset -- 中文指令微调数据集☆199Updated 7 months ago