stanfordnlp/pyreft

🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

Post-training with Tinker

Code for the paper "Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer"

A library for mechanistic interpretability of GPT-style language models

Train transformer language models with reinforcement learning.

Implementation of the training framework proposed in Self-Rewarding Language Model, from MetaAI

Bringing BERT into modernity via both architecture changes and scaling

PyTorch native post-training library

A Lightweight LLM Post-Training Library