Automatic Feature Extraction Using Deep Neural Models

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.

A 13B large language model developed by Baichuan Intelligent Technology

A large-scale 7B pretraining language model developed by BaiChuan-Inc.

A series of large language models developed by Baichuan Intelligent Technology

Efficient GPU kernels for block-sparse matrix multiplication and convolution

SoTA Transformers with C-backend for fast inference on your CPU.

C++ implementation of ChatGLM-6B & ChatGLM2-6B & ChatGLM3 & more LLMs

ChatGLM3 series: Open Bilingual Chat LLMs | 开源双语对话语言模型

ChatPDF is a Streamlit app allowing users to query PDF & DOCX content via natural language. It indexes documents for conversational interactions using OpenAI's API & FAISS for rapid matching. Uploa…

中文LLaMA-2 & Alpaca-2大模型二期项目 + 16K超长上下文模型 (Chinese LLaMA-2 & Alpaca-2 LLMs, including 16K long context models)

Chinese LLaMA2 - Enhanced LLaMA2 with PT/SFT/RLHF

CodeGeeX2: A More Powerful Multilingual Code Generation Model

Inference code for CodeLlama models

A Library for Dynamic Graph Learning (NeurIPS 2023)

FinGLM: 致力于构建一个开放的、公益的、持久的金融大模型项目，利用开源开放来促进「AI+金融」。

Fast and memory-efficient exact attention

Tensor library for machine learning

Simple and efficient pytorch-native transformer text generation in <1000 LOC of python.

this repository accompanies the book "Grokking Deep Learning"

Code and data for the KDD 2023 paper "Predicting Information Pathways Across Online Communities."

A PyTorch implementation of ACM SIGKDD 2019 paper "Predicting Dynamic Embedding Trajectory in Temporal Interaction Networks"

Langchain-Chatchat（原Langchain-ChatGLM）基于 Langchain 与 ChatGLM 等语言模型的本地知识库问答 | Langchain-Chatchat (formerly langchain-ChatGLM), local knowledge based LLM (like ChatGLM) QA app with langchain

Implementation of the LLaMA language model based on nanoGPT. Supports flash attention, Int8 and GPTQ 4bit quantization, LoRA and LLaMA-Adapter fine-tuning, pre-training. Apache 2.0-licensed.

Inference code for LLaMA models

Progressive LLaMA with Block Expansion.

Port of Facebook's LLaMA model in C/C++