12 Weeks, 24 Lessons, AI for All!

《开源大模型食用指南》基于Linux环境快速部署开源大模型，更适合中国宝宝的部署教程

Benchmarking large language models' complex reasoning ability with chain-of-thought prompting

BERT score for text generation

Official implementation for "Automatic Chain of Thought Prompting in Large Language Models" (stay tuned & more will be updated)

HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal

Code release for REPAIR: REnormalizing Permuted Activations for Interpolation Repair

SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusal Behaviors

analytic solution to the git-merge algorithm, derived from "Git Re-Basin: Merging Models modulo Permutation Symmetries"

Assessing Safety and Robustness of LLM's by performing Adversarial attacks.