{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:VIRGBTQSNNUQIMDVGANX2IRHEA","short_pith_number":"pith:VIRGBTQS","canonical_record":{"source":{"id":"2403.17806","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-03-26T15:44:58Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"c1f99394efce96d99df7e1d42f24f1e85e1614ced57328de4db2230c24d17cf6","abstract_canon_sha256":"ba2dd3fab323aab38bd3e1d8d8442edf1ed68cdde3a026609b15954809bc2d68"},"schema_version":"1.0"},"canonical_sha256":"aa2260ce126b69043075301b7d2227201c7cff72f5c181f9e61d48b4f97a233a","source":{"kind":"arxiv","id":"2403.17806","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.17806","created_at":"2026-07-05T08:43:54Z"},{"alias_kind":"arxiv_version","alias_value":"2403.17806v2","created_at":"2026-07-05T08:43:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.17806","created_at":"2026-07-05T08:43:54Z"},{"alias_kind":"pith_short_12","alias_value":"VIRGBTQSNNUQ","created_at":"2026-07-05T08:43:54Z"},{"alias_kind":"pith_short_16","alias_value":"VIRGBTQSNNUQIMDV","created_at":"2026-07-05T08:43:54Z"},{"alias_kind":"pith_short_8","alias_value":"VIRGBTQS","created_at":"2026-07-05T08:43:54Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:VIRGBTQSNNUQIMDVGANX2IRHEA","target":"record","payload":{"canonical_record":{"source":{"id":"2403.17806","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-03-26T15:44:58Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"c1f99394efce96d99df7e1d42f24f1e85e1614ced57328de4db2230c24d17cf6","abstract_canon_sha256":"ba2dd3fab323aab38bd3e1d8d8442edf1ed68cdde3a026609b15954809bc2d68"},"schema_version":"1.0"},"canonical_sha256":"aa2260ce126b69043075301b7d2227201c7cff72f5c181f9e61d48b4f97a233a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:43:54.796812Z","signature_b64":"RBcy7Kh+/tRYwf9tbAMcytDDRGgxD2EH8pDintu6ZchxOBcPadpPWXK7fVaznuY2ENCKsDcplhPsNmHl8ccSCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"aa2260ce126b69043075301b7d2227201c7cff72f5c181f9e61d48b4f97a233a","last_reissued_at":"2026-07-05T08:43:54.796395Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:43:54.796395Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2403.17806","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":"2026-07-05T08:43:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"eZrTEgnOf2lFAV60BMjTXWtldwunAVsZhCAce8eD/7CsQT09kiHKim+YHTv1ogxrZ1Art8nRWAQz8ZaiavdiDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T03:34:19.961909Z"},"content_sha256":"23ae218389e87492965bbb4e9806b8f790a672e6ea2fcd0cac01878791c904a6","schema_version":"1.0","event_id":"sha256:23ae218389e87492965bbb4e9806b8f790a672e6ea2fcd0cac01878791c904a6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:VIRGBTQSNNUQIMDVGANX2IRHEA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Have Faith in Faithfulness: Going Beyond Circuit Overlap When Finding Model Mechanisms","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Michael Hanna, Sandro Pezzelle, Yonatan Belinkov","submitted_at":"2024-03-26T15:44:58Z","abstract_excerpt":"Many recent language model (LM) interpretability studies have adopted the circuits framework, which aims to find the minimal computational subgraph, or circuit, that explains LM behavior on a given task. Most studies determine which edges belong in a LM's circuit by performing causal interventions on each edge independently, but this scales poorly with model size. Edge attribution patching (EAP), gradient-based approximation to interventions, has emerged as a scalable but imperfect solution to this problem. In this paper, we introduce a new method - EAP with integrated gradients (EAP-IG) - tha"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.17806","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/2403.17806/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":"2026-07-05T08:43:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZS/HUujs8nMu+fZopWisQX3/xGZfHUqhDOx2EN0e1LNZr4mu9gB94wnV30/fn0u2EEwZZOVZwCk1frNZdUGGBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T03:34:19.962460Z"},"content_sha256":"75e8ab5572630bdd4ca996f213ad7ad156010f094827eecf0a3475c34a341051","schema_version":"1.0","event_id":"sha256:75e8ab5572630bdd4ca996f213ad7ad156010f094827eecf0a3475c34a341051"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/VIRGBTQSNNUQIMDVGANX2IRHEA/bundle.json","state_url":"https://pith.science/pith/VIRGBTQSNNUQIMDVGANX2IRHEA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/VIRGBTQSNNUQIMDVGANX2IRHEA/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":"2026-08-07T03:34:19Z","links":{"resolver":"https://pith.science/pith/VIRGBTQSNNUQIMDVGANX2IRHEA","bundle":"https://pith.science/pith/VIRGBTQSNNUQIMDVGANX2IRHEA/bundle.json","state":"https://pith.science/pith/VIRGBTQSNNUQIMDVGANX2IRHEA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/VIRGBTQSNNUQIMDVGANX2IRHEA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:VIRGBTQSNNUQIMDVGANX2IRHEA","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":"ba2dd3fab323aab38bd3e1d8d8442edf1ed68cdde3a026609b15954809bc2d68","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-03-26T15:44:58Z","title_canon_sha256":"c1f99394efce96d99df7e1d42f24f1e85e1614ced57328de4db2230c24d17cf6"},"schema_version":"1.0","source":{"id":"2403.17806","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.17806","created_at":"2026-07-05T08:43:54Z"},{"alias_kind":"arxiv_version","alias_value":"2403.17806v2","created_at":"2026-07-05T08:43:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.17806","created_at":"2026-07-05T08:43:54Z"},{"alias_kind":"pith_short_12","alias_value":"VIRGBTQSNNUQ","created_at":"2026-07-05T08:43:54Z"},{"alias_kind":"pith_short_16","alias_value":"VIRGBTQSNNUQIMDV","created_at":"2026-07-05T08:43:54Z"},{"alias_kind":"pith_short_8","alias_value":"VIRGBTQS","created_at":"2026-07-05T08:43:54Z"}],"graph_snapshots":[{"event_id":"sha256:75e8ab5572630bdd4ca996f213ad7ad156010f094827eecf0a3475c34a341051","target":"graph","created_at":"2026-07-05T08:43:54Z","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/2403.17806/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Many recent language model (LM) interpretability studies have adopted the circuits framework, which aims to find the minimal computational subgraph, or circuit, that explains LM behavior on a given task. Most studies determine which edges belong in a LM's circuit by performing causal interventions on each edge independently, but this scales poorly with model size. Edge attribution patching (EAP), gradient-based approximation to interventions, has emerged as a scalable but imperfect solution to this problem. In this paper, we introduce a new method - EAP with integrated gradients (EAP-IG) - tha","authors_text":"Michael Hanna, Sandro Pezzelle, Yonatan Belinkov","cross_cats":["cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-03-26T15:44:58Z","title":"Have Faith in Faithfulness: Going Beyond Circuit Overlap When Finding Model Mechanisms"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.17806","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":"sha256:23ae218389e87492965bbb4e9806b8f790a672e6ea2fcd0cac01878791c904a6","target":"record","created_at":"2026-07-05T08:43:54Z","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":"ba2dd3fab323aab38bd3e1d8d8442edf1ed68cdde3a026609b15954809bc2d68","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-03-26T15:44:58Z","title_canon_sha256":"c1f99394efce96d99df7e1d42f24f1e85e1614ced57328de4db2230c24d17cf6"},"schema_version":"1.0","source":{"id":"2403.17806","kind":"arxiv","version":2}},"canonical_sha256":"aa2260ce126b69043075301b7d2227201c7cff72f5c181f9e61d48b4f97a233a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"aa2260ce126b69043075301b7d2227201c7cff72f5c181f9e61d48b4f97a233a","first_computed_at":"2026-07-05T08:43:54.796395Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:43:54.796395Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"RBcy7Kh+/tRYwf9tbAMcytDDRGgxD2EH8pDintu6ZchxOBcPadpPWXK7fVaznuY2ENCKsDcplhPsNmHl8ccSCg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:43:54.796812Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.17806","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:23ae218389e87492965bbb4e9806b8f790a672e6ea2fcd0cac01878791c904a6","sha256:75e8ab5572630bdd4ca996f213ad7ad156010f094827eecf0a3475c34a341051"],"state_sha256":"ee27d2a75db0948d263d675b0429cf012f52260ed565949fdc55a843324001fa"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/TDjBekZKmi4v/nLouvyUEHuZjWj4lYItKSTD6aPWDKn1R75DlE81r1xq9xpJ8v6lTN4gEX47xYJaSD4ao1zDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T03:34:19.967694Z","bundle_sha256":"4335f081ad3ce00c927bc6f233f49718a13e5915a797d124ac047efe2507f56c"}}