Sparse Auto-Encoders (SAEs) for Mechanistic Interpretability Collection A compilation of sparse auto-encoders trained on large language models. โข 37 items โข Updated 12 days ago โข 16
Running 89 The Eiffel Tower Llama ๐ 89 Explore the Eiffel Tower Llama experiment with open-source models
Running on Zero 7 The Eiffel Tower Llama Demo ๐ฌ 7 Steering a large language model using sparse autoencoders
Running on Zero 7 The Eiffel Tower Llama Demo ๐ฌ 7 Steering a large language model using sparse autoencoders
Running 89 The Eiffel Tower Llama ๐ 89 Explore the Eiffel Tower Llama experiment with open-source models
Running on Zero 7 The Eiffel Tower Llama Demo ๐ฌ 7 Steering a large language model using sparse autoencoders