{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/llama-scope-extracting-millions-of-features","title":"Llama Scope: Extracting Millions of Features from Llama-3.1-8B with Sparse Autoencoders","arxiv_id":"2410.20526","date":"2024-10-27","proceeding":null,"authors":["Zhengfu He","Wentao Shu","Xuyang Ge","Lingjie Chen","Junxuan Wang","Yunhua Zhou","Frances Liu","Qipeng Guo","Xuanjing Huang","Zuxuan Wu","Yu-Gang Jiang","Xipeng Qiu"],"abstract":"Sparse Autoencoders (SAEs) have emerged as a powerful unsupervised method for extracting sparse representations from language models, yet scalable training remains a significant challenge. We introduce a suite of 256 SAEs, trained on each layer and sublayer of the Llama-3.1-8B-Base model, with 32K and 128K features. Modifications to a state-of-the-art SAE variant, Top-K SAEs, are evaluated across multiple dimensions. In particular, we assess the generalizability of SAEs trained on base models to longer contexts and fine-tuned models. Additionally, we analyze the geometry of learned SAE latents, confirming that \\emph{feature splitting} enables the discovery of new features. The Llama Scope SAE checkpoints are publicly available at~\\url{https://huggingface.co/fnlp/Llama-Scope}, alongside our scalable training, interpretation, and visualization tools at \\url{https://github.com/OpenMOSS/Language-Model-SAEs}. These contributions aim to advance the open-source Sparse Autoencoder ecosystem and support mechanistic interpretability research by reducing the need for redundant SAE training.","url_abs":"https://arxiv.org/abs/2410.20526v1","url_pdf":"https://arxiv.org/pdf/2410.20526v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"llama-scope-extracting-millions-of-features","repo_url":"https://github.com/openmoss/language-model-saes","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"jax","reach":{"status":"ok"}}],"tasks":[{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"}],"methods":[{"method_slug":"base","method_name":"BASE"},{"method_slug":"llama","method_name":"LLaMA"},{"method_slug":"sparse-autoencoder","method_name":"Sparse Autoencoder"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2410.20526","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2410.20526"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/openmoss/language-model-saes","reach":{"status":"ok"}}],"summary":{"unverified":4},"by_repo_kind":{"official":{"samples":4,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":4,"samples":[{"code_sha256_prefix":"9c10445c3187eedd","entry":"get_scheduler","repo":"openmoss/language-model-saes","repo_kind":"official","path":"src/llamascopium/optim.py","file_url":"https://github.com/openmoss/language-model-saes/blob/HEAD/src/llamascopium/optim.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"9c10445c3187eedd"}},{"code_sha256_prefix":"df3038eee6139723","entry":"infer_model_backend","repo":"openmoss/language-model-saes","repo_kind":"official","path":"src/llamascopium/resource_loaders.py","file_url":"https://github.com/openmoss/language-model-saes/blob/HEAD/src/llamascopium/resource_loaders.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"df3038eee6139723"}},{"code_sha256_prefix":"4cef3a5f85d0fa05","entry":"set_tokens","repo":"openmoss/language-model-saes","repo_kind":"official","path":"src/llamascopium/backend/language_model.py","file_url":"https://github.com/openmoss/language-model-saes/blob/HEAD/src/llamascopium/backend/language_model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"4cef3a5f85d0fa05"}},{"code_sha256_prefix":"62244a8d08417415","entry":"to_tokens","repo":"openmoss/language-model-saes","repo_kind":"official","path":"src/llamascopium/backend/language_model.py","file_url":"https://github.com/openmoss/language-model-saes/blob/HEAD/src/llamascopium/backend/language_model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"62244a8d08417415"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}