wei-potato / Train-llm-from-scratch
使用deepspeed从头开始训练一个LLM,经过pretrain和sft阶段,验证llm学习知识、理解语言、回答问题的能力
☆152Updated 9 months ago
Alternatives and similar repositories for Train-llm-from-scratch:
Users that are interested in Train-llm-from-scratch are comparing it to the libraries listed below
- This is a repo for my NanoGPT Pytorch2.0 Implementation when torch2.0 released soon, faster and simpler, a good tutorial learning GPT.☆52Updated last year
- ☆121Updated last month
- Support mixed-precsion inference with vllm☆83Updated 3 months ago
- 用VLLM框架部署千问1.5并进行流式输出☆91Updated last year
- Mixed precision inference by Tensorrt-LLM☆79Updated 6 months ago
- [ACL 2024] User-friendly evaluation framework: Eval Suite & Benchmarks: UHGEval, HaluEval, HalluQA, etc.☆162Updated 5 months ago
- The framework to prune LLMs to any size and any config.☆91Updated last year
- This tool(enhance_long) aims to enhance the LlaMa2 long context extrapolation capability in the lowest-cost approach, preferably without …☆45Updated last year
- ☆44Updated 3 weeks ago
- Chinese large language model☆118Updated last year
- DeepRetrieval - Hacking 🔥Real Search Engines and Retrievers with LLM via RL☆423Updated last week
- 从预训练到强化学习的中文llama2☆88Updated last year
- Emotion text classification using Llama3-8b with LoRA and FlashAttention. Based on LLaMA-Factory.☆66Updated 8 months ago
- Higher performance OpenAI LLM service than vLLM serve: A pure C++ high-performance OpenAI LLM service implemented with GPRS+TensorRT-LLM+…☆130Updated last week
- A collection of papers related to knowledge fusion☆54Updated 6 months ago
- ✨ A synthetic dataset generation framework that produces diverse coding questions and verifiable solutions - all in one framwork☆202Updated 3 weeks ago
- MPLSandbox is an out-of-the-box multi-programming language sandbox designed to provide unified and comprehensive feedback from compiler a…☆174Updated last week
- Official implementation of RARE: Retrieval-Augmented Reasoning Modeling [Work in Progress]☆77Updated last week
- Reverse Chain-of-Thought Problem Generation for Geometric Reasoning in Large Multimodal Models☆174Updated 5 months ago
- 本项目旨在结合以往研究人员的代表性工作，从多个维度评估sft数据，并自动化过滤sft数据。☆44Updated last year
- [NeurIPS 2024] BAdam: A Memory Efficient Full Parameter Optimization Method for Large Language Models☆252Updated last month
- 教你只用最基本的python语法和numpy一步步实现深度学习框架☆127Updated 8 months ago
- We leverage 14 datasets as OOD test data and conduct evaluations on 8 NLU tasks over 21 popularly used models. Our findings confirm that …☆94Updated last year
- Explore concepts like Self-Correct, Self-Refine, Self-Improve, Self-Contradict, Self-Play, and Self-Knowledge, alongside o1-like reasonin…☆166Updated 4 months ago
- ✨✨latest advancements in VLA models(VIsion Language Action)☆56Updated last week
- adds Sequence Parallelism into LLaMA-Factory☆464Updated last week
- This includes the original implementation of CtrlA: Adaptive Retrieval-Augmented Generation via Inherent Control.☆60Updated 6 months ago
- Official Implementation of "Pay Attention to What You Need"☆42Updated 2 months ago
- An acceleration library that supports arbitrary bit-width combinatorial quantization operations☆222Updated 6 months ago
- Multilingual Corpus of Web Fiction☆193Updated 9 months ago