A Python implementation of John Gruber’s Markdown with Extension support.

PyTorch implementation of FractalGen https://arxiv.org/abs/2502.17437

This repo implements the Adam + SPD (selective projection decay) regularization.

The code repository for the CURLoRA research paper. Stable LLM continual fine-tuning and catastrophic forgetting mitigation.

Bayesian Parameter-Efficient Fine-Tuning for Overcoming Catastrophic Forgetting

A python package to simulate typographical errors.

800,000 step-level correctness labels on LLM solutions to MATH problems

An extremely fast Python package and project manager, written in Rust.

Clean, minimal, accessible reproduction of DeepSeek R1-Zero

Aligning pretrained language models with instruction data generated by themselves.

tiktoken is a fast BPE tokeniser for use with OpenAI's models.

Scalene: a high-performance, high-precision CPU, GPU, and memory profiler for Python with AI-powered optimization proposals

List of AI Residency Programs

Code for paper: Aligning Large Language Models with Representation Editing: A Control Perspective

Yet another alternative curriculum vitae/résumé class with LaTeX

Skywork Reward Model Series

Tomorrow Theme

A Terminal theme that mimics the One Dark theme made by the Atom team.

A Python module for creating Excel XLSX files.

RewardBench: the first evaluation tool for reward models.

Rich is a Python library for rich text and beautiful formatting in the terminal.

The Levenshtein Python C extension module contains functions for fast computation of Levenshtein distance and string similarity

Python composable command line interface toolkit

Simple cross-platform colored terminal text in Python

[MLSys 2024 Best Paper Award] AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration

OpenChat: Advancing Open-source Language Models with Imperfect Data

An open platform for training, serving, and evaluating large language models. Release repo for Vicuna and Chatbot Arena.

[ACL2023] We introduce LLM-Blender, an innovative ensembling framework to attain consistently superior performance by leveraging the diverse strengths of multiple open-source LLMs. LLM-Blender cut …