Stars
基于stablebaseline3强化学习框架和gym-super-mario-bros马里奥游戏包，训练马里奥通关。
🚀🚀 「大模型」2小时完全从0训练26M的小参数GPT！🌏 Train a 26M-parameter GPT from scratch in just 2h!
Official Implementation of "Neural Image Compression with Text-guided Encoding for both Pixel-level and Perceptual Fidelity (ICML 2024)"
A collection of papers related to data compression
📚A curated list of Awesome LLM/VLM Inference Papers with codes: WINT8/4, FlashAttention, PagedAttention, Parallelism, MLA, etc.
本项目旨在分享大模型相关技术原理以及实战经验（大模型工程化、大模型应用落地）
Features and Quality Dataset for DASH
用于从头预训练+SFT一个小参数量的中文LLaMa2的仓库；24G单卡即可运行得到一个具备简单中文问答能力的chat-llama2.