{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:3GMUMZVQSEE3KDQUZVRW3GE5DP","short_pith_number":"pith:3GMUMZVQ","schema_version":"1.0","canonical_sha256":"d9994666b09109b50e14cd636d989d1bff4ff13cfc8ef9f319439cafb8ae0c1e","source":{"kind":"arxiv","id":"2312.11560","version":3},"attestation_state":"computed","paper":{"title":"Learning from Emergence: A Study on Proactively Inhibiting the Monosemantic Neurons of Artificial Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.NE"],"primary_cat":"cs.LG","authors_text":"Charles Wang Wai Ng, Jiachuan Wang, Lei Chen, Shimin Di","submitted_at":"2023-12-17T14:42:46Z","abstract_excerpt":"Recently, emergence has received widespread attention from the research community along with the success of large-scale models. Different from the literature, we hypothesize a key factor that promotes the performance during the increase of scale: the reduction of monosemantic neurons that can only form one-to-one correlations with specific features. Monosemantic neurons tend to be sparser and have negative impacts on the performance in large models. Inspired by this insight, we propose an intuitive idea to identify monosemantic neurons and inhibit them. However, achieving this goal is a non-tr"},"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":"2312.11560","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-12-17T14:42:46Z","cross_cats_sorted":["cs.AI","cs.NE"],"title_canon_sha256":"f712e94d0fde53e7c2244cf4a6b84053ce37b0bbfcad3eaa080fbb383a12c592","abstract_canon_sha256":"7d61ea9cc0c18f5cd76ba9aabc5232ca3588e45756f27308460a409cba210244"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:34:09.080701Z","signature_b64":"uXlAeXrEy0rIefXTqiiQDACHY/rBK5PBOeztFqNWSYThJRz/LH9cS7eIlc3vpEMKw6EzOgN7O33HXeStgG3qAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d9994666b09109b50e14cd636d989d1bff4ff13cfc8ef9f319439cafb8ae0c1e","last_reissued_at":"2026-07-05T08:34:09.080159Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:34:09.080159Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning from Emergence: A Study on Proactively Inhibiting the Monosemantic Neurons of Artificial Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.NE"],"primary_cat":"cs.LG","authors_text":"Charles Wang Wai Ng, Jiachuan Wang, Lei Chen, Shimin Di","submitted_at":"2023-12-17T14:42:46Z","abstract_excerpt":"Recently, emergence has received widespread attention from the research community along with the success of large-scale models. Different from the literature, we hypothesize a key factor that promotes the performance during the increase of scale: the reduction of monosemantic neurons that can only form one-to-one correlations with specific features. Monosemantic neurons tend to be sparser and have negative impacts on the performance in large models. Inspired by this insight, we propose an intuitive idea to identify monosemantic neurons and inhibit them. However, achieving this goal is a non-tr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.11560","kind":"arxiv","version":3},"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/2312.11560/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":"2312.11560","created_at":"2026-07-05T08:34:09.080219+00:00"},{"alias_kind":"arxiv_version","alias_value":"2312.11560v3","created_at":"2026-07-05T08:34:09.080219+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.11560","created_at":"2026-07-05T08:34:09.080219+00:00"},{"alias_kind":"pith_short_12","alias_value":"3GMUMZVQSEE3","created_at":"2026-07-05T08:34:09.080219+00:00"},{"alias_kind":"pith_short_16","alias_value":"3GMUMZVQSEE3KDQU","created_at":"2026-07-05T08:34:09.080219+00:00"},{"alias_kind":"pith_short_8","alias_value":"3GMUMZVQ","created_at":"2026-07-05T08:34:09.080219+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.16374","citing_title":"SAFR: Neuron Redistribution for Interpretability","ref_index":26,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3GMUMZVQSEE3KDQUZVRW3GE5DP","json":"https://pith.science/pith/3GMUMZVQSEE3KDQUZVRW3GE5DP.json","graph_json":"https://pith.science/api/pith-number/3GMUMZVQSEE3KDQUZVRW3GE5DP/graph.json","events_json":"https://pith.science/api/pith-number/3GMUMZVQSEE3KDQUZVRW3GE5DP/events.json","paper":"https://pith.science/paper/3GMUMZVQ"},"agent_actions":{"view_html":"https://pith.science/pith/3GMUMZVQSEE3KDQUZVRW3GE5DP","download_json":"https://pith.science/pith/3GMUMZVQSEE3KDQUZVRW3GE5DP.json","view_paper":"https://pith.science/paper/3GMUMZVQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2312.11560&json=true","fetch_graph":"https://pith.science/api/pith-number/3GMUMZVQSEE3KDQUZVRW3GE5DP/graph.json","fetch_events":"https://pith.science/api/pith-number/3GMUMZVQSEE3KDQUZVRW3GE5DP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3GMUMZVQSEE3KDQUZVRW3GE5DP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3GMUMZVQSEE3KDQUZVRW3GE5DP/action/storage_attestation","attest_author":"https://pith.science/pith/3GMUMZVQSEE3KDQUZVRW3GE5DP/action/author_attestation","sign_citation":"https://pith.science/pith/3GMUMZVQSEE3KDQUZVRW3GE5DP/action/citation_signature","submit_replication":"https://pith.science/pith/3GMUMZVQSEE3KDQUZVRW3GE5DP/action/replication_record"}},"created_at":"2026-07-05T08:34:09.080219+00:00","updated_at":"2026-07-05T08:34:09.080219+00:00"}