{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:PMMBD5F6WLHZHSGOFKJDM34BAS","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":"8631b6c4d7152b88780b79a3be7f4bb883aa0abb05760b0548a401d613ae6602","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-14T17:11:48Z","title_canon_sha256":"57730b092d33486a823d553c0362fe9ea52840ed7858b1f7ca11696e5012ec24"},"schema_version":"1.0","source":{"id":"2506.12576","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.12576","created_at":"2026-07-05T11:28:37Z"},{"alias_kind":"arxiv_version","alias_value":"2506.12576v2","created_at":"2026-07-05T11:28:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.12576","created_at":"2026-07-05T11:28:37Z"},{"alias_kind":"pith_short_12","alias_value":"PMMBD5F6WLHZ","created_at":"2026-07-05T11:28:37Z"},{"alias_kind":"pith_short_16","alias_value":"PMMBD5F6WLHZHSGO","created_at":"2026-07-05T11:28:37Z"},{"alias_kind":"pith_short_8","alias_value":"PMMBD5F6","created_at":"2026-07-05T11:28:37Z"}],"graph_snapshots":[{"event_id":"sha256:5263e5bd89d372fb70b75bf6b970a4dff6296af85fb20e51cfde966972c68317","target":"graph","created_at":"2026-07-05T11:28:37Z","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/2506.12576/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent work shows that Sparse Autoencoders (SAE) applied to large language model (LLM) layers have neurons corresponding to interpretable concepts. These SAE neurons can be modified to align generated outputs, but only towards pre-identified topics and with some parameter tuning. Our approach leverages the observational and modification properties of SAEs to enable alignment for any topic. This method 1) scores each SAE neuron by its semantic similarity to an alignment text and uses them to 2) modify SAE-layer-level outputs by emphasizing topic-aligned neurons. We assess the alignment capabili","authors_text":"Ananya Joshi, Celia Cintas, Skyler Speakman","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-14T17:11:48Z","title":"Enabling Precise Topic Alignment in Large Language Models Via Sparse Autoencoders"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.12576","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":"sha256:7dfa3fb84e2f27850ea19cc3ed7136bbe5bf9656235997e146b7d7f2312b5ec9","target":"record","created_at":"2026-07-05T11:28:37Z","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":"8631b6c4d7152b88780b79a3be7f4bb883aa0abb05760b0548a401d613ae6602","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-14T17:11:48Z","title_canon_sha256":"57730b092d33486a823d553c0362fe9ea52840ed7858b1f7ca11696e5012ec24"},"schema_version":"1.0","source":{"id":"2506.12576","kind":"arxiv","version":2}},"canonical_sha256":"7b1811f4beb2cf93c8ce2a92366f810497c78cd5c4d1305349033eae0574896d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7b1811f4beb2cf93c8ce2a92366f810497c78cd5c4d1305349033eae0574896d","first_computed_at":"2026-07-05T11:28:37.525836Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:28:37.525836Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"faK8a1TkaDlNm3BjHZ2ePKn1MTry6FK5GCNbFn2OcL51y35Zxbd5FUyffXKV8s/YFrInaXDxdUfQOgSSjluiDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:28:37.526349Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.12576","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7dfa3fb84e2f27850ea19cc3ed7136bbe5bf9656235997e146b7d7f2312b5ec9","sha256:5263e5bd89d372fb70b75bf6b970a4dff6296af85fb20e51cfde966972c68317"],"state_sha256":"5c911f8d7efc9e97750c604d2486c5c6dc5310e5f4404265f52147dba9c55623"}