eric-mitchell

direct-preference-optimization

Reference implementation for DPO (Direct Preference Optimization)

detect-gpt

DetectGPT: Zero-Shot Machine-Generated Text Detection using Probability Curvature

mend

MEND: Fast Model Editing at Scale

serac

Semi-Parametric Editing with a Retrieval-Augmented Counterfactual Model

macaw

Code for "Offline Meta-Reinforcement Learning with Advantage Weighting" [ICML 2021]

macaw-min

Clean, extensible implementation of MACAW [ICML 2021]

concord

emulated-fine-tuning

An Emulator for Fine-tuning Large Language Models using Small Language Models

eric-mitchell.github.io

akima

Article extraction and annotation.

avicenna-theme

base_project

Bare project template with basic Guild/Torch config

dart

Deep Autoregressive Tensor Trains

disentanglement

higher

higher is a pytorch library allowing users to obtain higher order gradients over losses spanning training loops rather than individual training steps.

higher_selfdestruct

KnowledgeEditor

Code for Editing Factual Knowledge in Language Models

metaworld

oyster

Implementation of Efficient Off-policy Meta-learning via Probabilistic Context Variables (PEARL)

transformers

🤗Transformers: State-of-the-art Natural Language Processing for Pytorch and TensorFlow 2.0.

world-models

Reimplementation of World-Models (Ha and Schmidhuber 2018) in pytorch