{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:VQNMFQ6AMLSXEZ7MVMNKZT356O","short_pith_number":"pith:VQNMFQ6A","canonical_record":{"source":{"id":"2511.16309","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-11-20T12:37:54Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"b81fa04cf4346eb3945e90691ad49d162a073575b4fa9ae4af1f3bba9036c7a6","abstract_canon_sha256":"c805adf1dc522d2110f058776bd322493c0afb33b17d7519c8e928689b145017"},"schema_version":"1.0"},"canonical_sha256":"ac1ac2c3c062e57267ecab1aaccf7df3928d90aa50772ab35230d1fc5702f69b","source":{"kind":"arxiv","id":"2511.16309","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2511.16309","created_at":"2026-05-20T00:04:17Z"},{"alias_kind":"arxiv_version","alias_value":"2511.16309v2","created_at":"2026-05-20T00:04:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2511.16309","created_at":"2026-05-20T00:04:17Z"},{"alias_kind":"pith_short_12","alias_value":"VQNMFQ6AMLSX","created_at":"2026-05-20T00:04:17Z"},{"alias_kind":"pith_short_16","alias_value":"VQNMFQ6AMLSXEZ7M","created_at":"2026-05-20T00:04:17Z"},{"alias_kind":"pith_short_8","alias_value":"VQNMFQ6A","created_at":"2026-05-20T00:04:17Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:VQNMFQ6AMLSXEZ7MVMNKZT356O","target":"record","payload":{"canonical_record":{"source":{"id":"2511.16309","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-11-20T12:37:54Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"b81fa04cf4346eb3945e90691ad49d162a073575b4fa9ae4af1f3bba9036c7a6","abstract_canon_sha256":"c805adf1dc522d2110f058776bd322493c0afb33b17d7519c8e928689b145017"},"schema_version":"1.0"},"canonical_sha256":"ac1ac2c3c062e57267ecab1aaccf7df3928d90aa50772ab35230d1fc5702f69b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:04:17.559667Z","signature_b64":"LFRlSEeiv8y9xnNhpGCToa1UKHvwiBVvRJ2HimQgFVV5D54A8IBcHql2Y5N2pzXuVzdQHe2+DA7w5H8bZZ+/Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ac1ac2c3c062e57267ecab1aaccf7df3928d90aa50772ab35230d1fc5702f69b","last_reissued_at":"2026-05-20T00:04:17.558757Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:04:17.558757Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2511.16309","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:04:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"++ocFKG38RGeVYy/Kajn3BzFoAaJK/B7zldFFsHZu1783f2L1s0rM/5ct5qqfWXVG+8JAT9mCOudRQAKSr1SAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-11T19:49:36.661496Z"},"content_sha256":"3c5b6ee96955520dbdb689b61045cc0c5ded3a638057578910a6e0e64bb21038","schema_version":"1.0","event_id":"sha256:3c5b6ee96955520dbdb689b61045cc0c5ded3a638057578910a6e0e64bb21038"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:VQNMFQ6AMLSXEZ7MVMNKZT356O","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Sparse Autoencoders are Topic Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Leander Girrbach, Zeynep Akata","submitted_at":"2025-11-20T12:37:54Z","abstract_excerpt":"Sparse autoencoders (SAEs) are used to analyze embeddings, but their role and practical value are debated. We propose a new perspective on SAEs by demonstrating that they can be naturally understood as topic models. We propose a continuous topic model (CTM) inspired by Latent Dirichlet Allocation (LDA) for embedding spaces and derive the SAE objective as a maximum a posteriori estimator under this model. This view implies SAE features are thematic components rather than steerable directions. To confirm our theoretical findings, we introduce SAE-TM, a topic modeling framework that: (1) trains a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2511.16309","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2511.16309/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:04:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pBoraey0DHkqJYGofA9D00PhOvJLn2Lda3en3E5U9bW8bqXaitd9d0qus1a14QaBhmD2UAil4u3KTEGCJSs4Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-11T19:49:36.662297Z"},"content_sha256":"2a7d6d357bded9f06854f5c7b46b8d750884b7493b48b9af49c9f4422ee0de56","schema_version":"1.0","event_id":"sha256:2a7d6d357bded9f06854f5c7b46b8d750884b7493b48b9af49c9f4422ee0de56"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/VQNMFQ6AMLSXEZ7MVMNKZT356O/bundle.json","state_url":"https://pith.science/pith/VQNMFQ6AMLSXEZ7MVMNKZT356O/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/VQNMFQ6AMLSXEZ7MVMNKZT356O/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-11T19:49:36Z","links":{"resolver":"https://pith.science/pith/VQNMFQ6AMLSXEZ7MVMNKZT356O","bundle":"https://pith.science/pith/VQNMFQ6AMLSXEZ7MVMNKZT356O/bundle.json","state":"https://pith.science/pith/VQNMFQ6AMLSXEZ7MVMNKZT356O/state.json","well_known_bundle":"https://pith.science/.well-known/pith/VQNMFQ6AMLSXEZ7MVMNKZT356O/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:VQNMFQ6AMLSXEZ7MVMNKZT356O","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"c805adf1dc522d2110f058776bd322493c0afb33b17d7519c8e928689b145017","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-11-20T12:37:54Z","title_canon_sha256":"b81fa04cf4346eb3945e90691ad49d162a073575b4fa9ae4af1f3bba9036c7a6"},"schema_version":"1.0","source":{"id":"2511.16309","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2511.16309","created_at":"2026-05-20T00:04:17Z"},{"alias_kind":"arxiv_version","alias_value":"2511.16309v2","created_at":"2026-05-20T00:04:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2511.16309","created_at":"2026-05-20T00:04:17Z"},{"alias_kind":"pith_short_12","alias_value":"VQNMFQ6AMLSX","created_at":"2026-05-20T00:04:17Z"},{"alias_kind":"pith_short_16","alias_value":"VQNMFQ6AMLSXEZ7M","created_at":"2026-05-20T00:04:17Z"},{"alias_kind":"pith_short_8","alias_value":"VQNMFQ6A","created_at":"2026-05-20T00:04:17Z"}],"graph_snapshots":[{"event_id":"sha256:2a7d6d357bded9f06854f5c7b46b8d750884b7493b48b9af49c9f4422ee0de56","target":"graph","created_at":"2026-05-20T00:04:17Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2511.16309/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Sparse autoencoders (SAEs) are used to analyze embeddings, but their role and practical value are debated. We propose a new perspective on SAEs by demonstrating that they can be naturally understood as topic models. We propose a continuous topic model (CTM) inspired by Latent Dirichlet Allocation (LDA) for embedding spaces and derive the SAE objective as a maximum a posteriori estimator under this model. This view implies SAE features are thematic components rather than steerable directions. To confirm our theoretical findings, we introduce SAE-TM, a topic modeling framework that: (1) trains a","authors_text":"Leander Girrbach, Zeynep Akata","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-11-20T12:37:54Z","title":"Sparse Autoencoders are Topic Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2511.16309","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:3c5b6ee96955520dbdb689b61045cc0c5ded3a638057578910a6e0e64bb21038","target":"record","created_at":"2026-05-20T00:04:17Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"c805adf1dc522d2110f058776bd322493c0afb33b17d7519c8e928689b145017","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-11-20T12:37:54Z","title_canon_sha256":"b81fa04cf4346eb3945e90691ad49d162a073575b4fa9ae4af1f3bba9036c7a6"},"schema_version":"1.0","source":{"id":"2511.16309","kind":"arxiv","version":2}},"canonical_sha256":"ac1ac2c3c062e57267ecab1aaccf7df3928d90aa50772ab35230d1fc5702f69b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ac1ac2c3c062e57267ecab1aaccf7df3928d90aa50772ab35230d1fc5702f69b","first_computed_at":"2026-05-20T00:04:17.558757Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:04:17.558757Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"LFRlSEeiv8y9xnNhpGCToa1UKHvwiBVvRJ2HimQgFVV5D54A8IBcHql2Y5N2pzXuVzdQHe2+DA7w5H8bZZ+/Aw==","signature_status":"signed_v1","signed_at":"2026-05-20T00:04:17.559667Z","signed_message":"canonical_sha256_bytes"},"source_id":"2511.16309","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3c5b6ee96955520dbdb689b61045cc0c5ded3a638057578910a6e0e64bb21038","sha256:2a7d6d357bded9f06854f5c7b46b8d750884b7493b48b9af49c9f4422ee0de56"],"state_sha256":"8d8c7283c9db06c7568050114d75d54b75494291ff850bc4a46c46b154ad90ea"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"eOnR52+iX2ddyJwo7MIBRbwUV3PXr/BVPU50Hstz30MfsO1qhtzgY25I1w/+Q2Kz9UYj5iyKeakbbw+mXHCVAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-11T19:49:36.666457Z","bundle_sha256":"1021872e32e630f9d8f847f4f1feedb1cd0917b6e36653d2eb92d1b891df6956"}}