• Towards Monosemanticity: Decomposing Language Models With Dictionary Learning

• **Sparse Autoencoders Find Highly Interpretable Features in Language Models (Logan)**

• Comparing Anthropic's Dictionary Learning to Ours

• [Interim research report] Taking features out of superposition with sparse autoencoders

• Sparse Autoencoders: Future Work

• Finding Neurons in a Haystack by Wes Gurnee

• Neel Nanda replication!!

PPO: https://huggingface.co/docs/trl/main/en/ppo_trainer

Upload Custom model to Hugging Face: https://huggingface.co/docs/transformers/custom_models#sharing-custom-models