Jbloo GPT2 SAE with Ghost Gradient Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small — LessWrongBrowse these SAE Features on Neuronpedia! …https://www.lesswrong.com/posts/f9EgfLSurAiqRJySD/open-source-sparse-autoencoders-for-all-residual-streamjbloom/GPT2-Small-SAEs-Reformatted at mainWe’re on a journey to advance and democratize artificial intelligence through open source and open science.https://huggingface.co/jbloom/GPT2-Small-SAEs-Reformatted/tree/mainDistributed trainingsparse_autoencoderopenai • Updated 2025 Aug 6 9:17 Attention SAEs for TGPT2Attention SAEs Scale to GPT-2 Small — LessWrongThis is an interim report that we are currently building on. We hope this update + open sourcing our SAEs will be useful to related research occurrin…https://www.lesswrong.com/posts/FSTRedtjuHa4Gfdbr/attention-saes-scale-to-gpt-2-smallckkissane/attn-saes-gpt2-small-all-layers · Hugging FaceWe’re on a journey to advance and democratize artificial intelligence through open source and open science.https://huggingface.co/ckkissane/attn-saes-gpt2-small-all-layersInduction head of GPT2We Inspected Every Head In GPT-2 Small using SAEs So You Don’t Have To — LessWrongThis is an interim report that we are currently building on. We hope this update will be useful to related research occurring in parallel. Produced a…https://www.lesswrong.com/posts/xmegeW5mqiBsvoaim/we-inspected-every-head-in-gpt-2-small-using-saes-so-you-don
Seonglae ChoSeonglae ChoSeonglae ChoSeonglae Cho
