{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:FCT5JLPDAX3M7VHEBC5VYUYR5I","short_pith_number":"pith:FCT5JLPD","schema_version":"1.0","canonical_sha256":"28a7d4ade305f6cfd4e408bb5c5311ea3e2ffb539c3d09436c3fe89425b106c6","source":{"kind":"arxiv","id":"2402.10688","version":2},"attestation_state":"computed","paper":{"title":"Towards Uncovering How Large Language Model Works: An Explainability Perspective","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bo Shen, Fan Yang, Haiyan Zhao, Himabindu Lakkaraju, Mengnan Du","submitted_at":"2024-02-16T13:46:06Z","abstract_excerpt":"Large language models (LLMs) have led to breakthroughs in language tasks, yet the internal mechanisms that enable their remarkable generalization and reasoning abilities remain opaque. This lack of transparency presents challenges such as hallucinations, toxicity, and misalignment with human values, hindering the safe and beneficial deployment of LLMs. This paper aims to uncover the mechanisms underlying LLM functionality through the lens of explainability. First, we review how knowledge is architecturally composed within LLMs and encoded in their internal parameters via mechanistic interpreta"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2402.10688","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-16T13:46:06Z","cross_cats_sorted":[],"title_canon_sha256":"e21c3cd1774e94b3db87bfda257e71badeaacf20b45c9e6f743f4c9af8e3ab6d","abstract_canon_sha256":"286d01901f7150e2466f26e7b2967fc00150884cca876171d2c4b63cf3d7fef7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:08:20.837678Z","signature_b64":"h7Pk2akSjYVr80HPvRrulaqN41iFWbqah7lAo/KfOMgT0gz3tOWodlxzACuFmv+L6Ncs9Lv3TGWrt5/PEtp3DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"28a7d4ade305f6cfd4e408bb5c5311ea3e2ffb539c3d09436c3fe89425b106c6","last_reissued_at":"2026-07-05T08:08:20.837231Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:08:20.837231Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Uncovering How Large Language Model Works: An Explainability Perspective","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bo Shen, Fan Yang, Haiyan Zhao, Himabindu Lakkaraju, Mengnan Du","submitted_at":"2024-02-16T13:46:06Z","abstract_excerpt":"Large language models (LLMs) have led to breakthroughs in language tasks, yet the internal mechanisms that enable their remarkable generalization and reasoning abilities remain opaque. This lack of transparency presents challenges such as hallucinations, toxicity, and misalignment with human values, hindering the safe and beneficial deployment of LLMs. This paper aims to uncover the mechanisms underlying LLM functionality through the lens of explainability. First, we review how knowledge is architecturally composed within LLMs and encoded in their internal parameters via mechanistic interpreta"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.10688","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/2402.10688/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2402.10688","created_at":"2026-07-05T08:08:20.837288+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.10688v2","created_at":"2026-07-05T08:08:20.837288+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.10688","created_at":"2026-07-05T08:08:20.837288+00:00"},{"alias_kind":"pith_short_12","alias_value":"FCT5JLPDAX3M","created_at":"2026-07-05T08:08:20.837288+00:00"},{"alias_kind":"pith_short_16","alias_value":"FCT5JLPDAX3M7VHE","created_at":"2026-07-05T08:08:20.837288+00:00"},{"alias_kind":"pith_short_8","alias_value":"FCT5JLPD","created_at":"2026-07-05T08:08:20.837288+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2511.12439","citing_title":"Multi-agent Self-triage System with Medical Flowcharts","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20467","citing_title":"Mechanistic Interpretability Tool for AI Weather Models","ref_index":18,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FCT5JLPDAX3M7VHEBC5VYUYR5I","json":"https://pith.science/pith/FCT5JLPDAX3M7VHEBC5VYUYR5I.json","graph_json":"https://pith.science/api/pith-number/FCT5JLPDAX3M7VHEBC5VYUYR5I/graph.json","events_json":"https://pith.science/api/pith-number/FCT5JLPDAX3M7VHEBC5VYUYR5I/events.json","paper":"https://pith.science/paper/FCT5JLPD"},"agent_actions":{"view_html":"https://pith.science/pith/FCT5JLPDAX3M7VHEBC5VYUYR5I","download_json":"https://pith.science/pith/FCT5JLPDAX3M7VHEBC5VYUYR5I.json","view_paper":"https://pith.science/paper/FCT5JLPD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.10688&json=true","fetch_graph":"https://pith.science/api/pith-number/FCT5JLPDAX3M7VHEBC5VYUYR5I/graph.json","fetch_events":"https://pith.science/api/pith-number/FCT5JLPDAX3M7VHEBC5VYUYR5I/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FCT5JLPDAX3M7VHEBC5VYUYR5I/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FCT5JLPDAX3M7VHEBC5VYUYR5I/action/storage_attestation","attest_author":"https://pith.science/pith/FCT5JLPDAX3M7VHEBC5VYUYR5I/action/author_attestation","sign_citation":"https://pith.science/pith/FCT5JLPDAX3M7VHEBC5VYUYR5I/action/citation_signature","submit_replication":"https://pith.science/pith/FCT5JLPDAX3M7VHEBC5VYUYR5I/action/replication_record"}},"created_at":"2026-07-05T08:08:20.837288+00:00","updated_at":"2026-07-05T08:08:20.837288+00:00"}