leauyn's repositories
baby-llama2-chinese
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.
Chatglm_lora_multi-gpu
chatglm多gpu用deepspeed和
Chinese-LLaMA-Alpaca
中文LLaMA&Alpaca大语言模型+本地CPU/GPU训练部署 (Chinese LLaMA & Alpaca LLMs)
Chinese-LLaMA-Alpaca-2
中文LLaMA-2 & Alpaca-2大模型二期项目 + 64K超长上下文模型 (Chinese LLaMA-2 & Alpaca-2 LLMs with 64K long context models)
cov-weighting
Implementation for our WACV 2021 paper "Multi-Loss Weighting with Coefficient of Variations"
Deep-Reinforcement-Learning-with-Python
Deep Reinforcement Learning with Python, Second Edition, published by Packt
DeepRec
推荐、广告工业界经典以及最前沿的论文、资料集合/ Must-read Papers on Recommendation System and CTR Prediction
DeepSpeedExamples
Example models using DeepSpeed
Firefly
Firefly: 大模型训练工具,支持训练MiniCPM、Yi、Deepseek、Orion、Xverse、Mixtral-8x7B、Zephyr、Mistral、Baichuan2、Llma2、Llama、Qwen、Baichuan、ChatGLM2、InternLM、Ziya2、Vicuna、Bloom等大模型
Firefly-LLaMA2-Chinese
Firefly中文LLaMA-2大模型,支持增量预训练Baichuan2、Llama2、Llama、Falcon、Qwen、Baichuan、InternLM、Bloom等大模型
flask
The Python micro framework for building web applications.
git_ex
Git Ex
how-to-train-tokenizer
怎么训练一个LLM分词器
hyperbolic-learning
Implemented ML algorithms in hyperbolic geometry (MDS, K-Means, Support vector machines, etc.)
hyperbolic_nn
Source code for the paper "Hyperbolic Neural Networks", https://arxiv.org/abs/1805.09112
Leetcode-retag
重新分类 Leetcode 高频题
llm-action
本项目旨在分享大模型相关技术原理以及实战经验。
llm-foundry
LLM training code for MosaicML foundation models
llm_interview_note
大模型面试题及答案,大模型八股文
Megatron-LM
Ongoing research training transformer models at scale
mixtral-offloading
Run Mixtral-8x7B models in Colab or consumer desktops
openwebtext
An open clone of the GPT-2 WebText dataset by OpenAI. Still WIP.
poincare_glove
Implementation of the "Poincare Glove: Hyperbolic word embeddings" paper
pytorch-a2c-ppo-acktr-gail
PyTorch implementation of Advantage Actor Critic (A2C), Proximal Policy Optimization (PPO), Scalable trust-region method for deep reinforcement learning using Kronecker-factored approximation (ACKTR) and Generative Adversarial Imitation Learning (GAIL).
simple_LLM_pretrain_learning_model
An example of quickly learning the basic principles and implementation of large models, based on lightweight data to complete the entire path of building large models. Gain a deep understanding of the process from theory to implementation of the Transformer, providing beginners with a fast entry path.
two-stream-action-recognition
My re-implementation of two stream action recognition