{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:6UZGADZMYHG6NVBEIAIFLUXH7X","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":"f8977356cb7d094d15c1bd682b0c120ac401f76d4709be723c7fc30fadc4acc1","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-29T15:33:39Z","title_canon_sha256":"9fb52cffbc396e5ae8968e55ce022d4154bc71c629fdd3e0b3f544dead9e2803"},"schema_version":"1.0","source":{"id":"2505.23556","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.23556","created_at":"2026-07-05T11:12:03Z"},{"alias_kind":"arxiv_version","alias_value":"2505.23556v1","created_at":"2026-07-05T11:12:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.23556","created_at":"2026-07-05T11:12:03Z"},{"alias_kind":"pith_short_12","alias_value":"6UZGADZMYHG6","created_at":"2026-07-05T11:12:03Z"},{"alias_kind":"pith_short_16","alias_value":"6UZGADZMYHG6NVBE","created_at":"2026-07-05T11:12:03Z"},{"alias_kind":"pith_short_8","alias_value":"6UZGADZM","created_at":"2026-07-05T11:12:03Z"}],"graph_snapshots":[{"event_id":"sha256:1fd02f31d4ad285fa834a63a101166eb5f691a2e1144e1a3ed04f47b8e97ff56","target":"graph","created_at":"2026-07-05T11:12:03Z","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/2505.23556/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Refusal is a key safety behavior in aligned language models, yet the internal mechanisms driving refusals remain opaque. In this work, we conduct a mechanistic study of refusal in instruction-tuned LLMs using sparse autoencoders to identify latent features that causally mediate refusal behaviors. We apply our method to two open-source chat models and intervene on refusal-related features to assess their influence on generation, validating their behavioral impact across multiple harmful datasets. This enables a fine-grained inspection of how refusal manifests at the activation level and address","authors_text":"Clement Neo, Erik Cambria, Nirmalendu Prakash, Ranjan Satapathy, Roy Ka-Wei Lee, Wei Jie Yeo","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-29T15:33:39Z","title":"Understanding Refusal in Language Models with Sparse Autoencoders"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.23556","kind":"arxiv","version":1},"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:d77bd42571d34e8f7ac40c2b5a2f76bc06be3723587948c014ceb919382616a6","target":"record","created_at":"2026-07-05T11:12:03Z","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":"f8977356cb7d094d15c1bd682b0c120ac401f76d4709be723c7fc30fadc4acc1","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-29T15:33:39Z","title_canon_sha256":"9fb52cffbc396e5ae8968e55ce022d4154bc71c629fdd3e0b3f544dead9e2803"},"schema_version":"1.0","source":{"id":"2505.23556","kind":"arxiv","version":1}},"canonical_sha256":"f532600f2cc1cde6d424401055d2e7fddaa2530b4ebbce14d5729e9f4a077a3a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f532600f2cc1cde6d424401055d2e7fddaa2530b4ebbce14d5729e9f4a077a3a","first_computed_at":"2026-07-05T11:12:03.862380Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:12:03.862380Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"cQ4A/DlGLzv42WgjIpAnYlxDp+3IkXqozE3TTrjc7YKlA0O0VEwr+NMctCKtchnEefVId9K2gk6Z/PioOv1lDg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:12:03.862859Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.23556","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d77bd42571d34e8f7ac40c2b5a2f76bc06be3723587948c014ceb919382616a6","sha256:1fd02f31d4ad285fa834a63a101166eb5f691a2e1144e1a3ed04f47b8e97ff56"],"state_sha256":"e5657858c5666412eb2f97095024d2a393ee6ab90a4bdaaba9f7fa73081d2852"}