The Gretel Python Client allows you to interact with the Gretel REST API.

Navigator Helpers

Aligning pretrained language models with instruction data generated by themselves.

Skywork series models are pre-trained on 3.2TB of high-quality multilingual (mainly Chinese and English) and code data. We have open-sourced the model, training data, evaluation data, evaluation me…

Code and documentation to train Stanford's Alpaca models, and generate the data.

Easily embed, cluster and semantically label text datasets

主要记录大语言大模型（LLMs） 算法（应用）工程师相关的知识及面试题

[ICML'24] The official implementation of “Rethinking Optimization and Architecture for Tiny Language Models”

Phi2-Chinese-0.2B 从0开始训练自己的Phi2中文小模型，支持接入langchain加载本地知识库做检索增强生成RAG。Training your own Phi2 small chat model from scratch.

Implementing a ChatGPT-like LLM in PyTorch from scratch, step by step

Fast and memory-efficient exact attention

The TinyLlama project is an open endeavor to pretrain a 1.1B Llama model on 3 trillion tokens.

DeepSeek-Coder-V2: Breaking the Barrier of Closed-Source Models in Code Intelligence

LiveBench: A Challenging, Contamination-Free LLM Benchmark

[ICLR 2024] Sheared LLaMA: Accelerating Language Model Pre-training via Structured Pruning

The hub for EleutherAI's work on interpretability and learning dynamics

The simplest, fastest repository for training/finetuning medium-sized GPTs.

DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models

Unofficial implementation of Google "CutPaste: Self-Supervised Learning for Anomaly Detection and Localization" in PyTorch

Repository for the paper "Rethinking Assumptions in Anomaly Detection"

Deep Anomaly Detection with Outlier Exposure (ICLR 2019)

official implementation of RoSAS: Deep Semi-supervised Anomaly Detection with Contamination-resilient Continuous Supervision

MiniCPM-2B: An end-side LLM outperforming Llama2-13B.

Jan is an open source alternative to ChatGPT that runs 100% offline on your computer. Multiple engine support (llama.cpp, TensorRT-LLM)

Tools for merging pretrained large language models.

Qwen2 is the large language model series developed by Qwen team, Alibaba Cloud.