{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:AOJHS7EFZDDFRPM7U2RXPSC2YI","short_pith_number":"pith:AOJHS7EF","schema_version":"1.0","canonical_sha256":"0392797c85c8c658bd9fa6a377c85ac20d519903af38cd4d3b1a87ca2e34ace4","source":{"kind":"arxiv","id":"2403.01509","version":2},"attestation_state":"computed","paper":{"title":"Fantastic Semantics and Where to Find Them: Investigating Which Layers of Generative LLMs Reflect Lexical Semantics","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Cunliang Kong, Maosong Sun, Ying Liu, Zhu Liu","submitted_at":"2024-03-03T13:14:47Z","abstract_excerpt":"Large language models have achieved remarkable success in general language understanding tasks. However, as a family of generative methods with the objective of next token prediction, the semantic evolution with the depth of these models are not fully explored, unlike their predecessors, such as BERT-like architectures. In this paper, we specifically investigate the bottom-up evolution of lexical semantics for a popular LLM, namely Llama2, by probing its hidden states at the end of each layer using a contextualized word identification task. Our experiments show that the representations in lowe"},"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":"2403.01509","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-03T13:14:47Z","cross_cats_sorted":[],"title_canon_sha256":"b19cae7d52adcffd51da91615ec72170adcd0bf0e044e08dc86c3b281b33dc59","abstract_canon_sha256":"42d58c7a7f5c530c35befacd5f41cf35e00e338e3457e598068c86ebf7aba24f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:29:23.553249Z","signature_b64":"GSDSScn27fwXBu6wuHmLwF61YUWMoMYJIRLJtMRevYeZ0yS/XE15skCFBrWXsAW6LvXwSrW5tKmZw/BqshCIDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0392797c85c8c658bd9fa6a377c85ac20d519903af38cd4d3b1a87ca2e34ace4","last_reissued_at":"2026-07-05T08:29:23.552831Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:29:23.552831Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Fantastic Semantics and Where to Find Them: Investigating Which Layers of Generative LLMs Reflect Lexical Semantics","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Cunliang Kong, Maosong Sun, Ying Liu, Zhu Liu","submitted_at":"2024-03-03T13:14:47Z","abstract_excerpt":"Large language models have achieved remarkable success in general language understanding tasks. However, as a family of generative methods with the objective of next token prediction, the semantic evolution with the depth of these models are not fully explored, unlike their predecessors, such as BERT-like architectures. In this paper, we specifically investigate the bottom-up evolution of lexical semantics for a popular LLM, namely Llama2, by probing its hidden states at the end of each layer using a contextualized word identification task. Our experiments show that the representations in lowe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.01509","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/2403.01509/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":"2403.01509","created_at":"2026-07-05T08:29:23.552886+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.01509v2","created_at":"2026-07-05T08:29:23.552886+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.01509","created_at":"2026-07-05T08:29:23.552886+00:00"},{"alias_kind":"pith_short_12","alias_value":"AOJHS7EFZDDF","created_at":"2026-07-05T08:29:23.552886+00:00"},{"alias_kind":"pith_short_16","alias_value":"AOJHS7EFZDDFRPM7","created_at":"2026-07-05T08:29:23.552886+00:00"},{"alias_kind":"pith_short_8","alias_value":"AOJHS7EF","created_at":"2026-07-05T08:29:23.552886+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.24956","citing_title":"NITP: Next Implicit Token Prediction for LLM Pre-training","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24956","citing_title":"NITP: Next Implicit Token Prediction for LLM Pre-training","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2505.11017","citing_title":"Logo-LLM: Local and Global Modeling with Large Language Models for Time Series Forecasting","ref_index":9,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AOJHS7EFZDDFRPM7U2RXPSC2YI","json":"https://pith.science/pith/AOJHS7EFZDDFRPM7U2RXPSC2YI.json","graph_json":"https://pith.science/api/pith-number/AOJHS7EFZDDFRPM7U2RXPSC2YI/graph.json","events_json":"https://pith.science/api/pith-number/AOJHS7EFZDDFRPM7U2RXPSC2YI/events.json","paper":"https://pith.science/paper/AOJHS7EF"},"agent_actions":{"view_html":"https://pith.science/pith/AOJHS7EFZDDFRPM7U2RXPSC2YI","download_json":"https://pith.science/pith/AOJHS7EFZDDFRPM7U2RXPSC2YI.json","view_paper":"https://pith.science/paper/AOJHS7EF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.01509&json=true","fetch_graph":"https://pith.science/api/pith-number/AOJHS7EFZDDFRPM7U2RXPSC2YI/graph.json","fetch_events":"https://pith.science/api/pith-number/AOJHS7EFZDDFRPM7U2RXPSC2YI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AOJHS7EFZDDFRPM7U2RXPSC2YI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AOJHS7EFZDDFRPM7U2RXPSC2YI/action/storage_attestation","attest_author":"https://pith.science/pith/AOJHS7EFZDDFRPM7U2RXPSC2YI/action/author_attestation","sign_citation":"https://pith.science/pith/AOJHS7EFZDDFRPM7U2RXPSC2YI/action/citation_signature","submit_replication":"https://pith.science/pith/AOJHS7EFZDDFRPM7U2RXPSC2YI/action/replication_record"}},"created_at":"2026-07-05T08:29:23.552886+00:00","updated_at":"2026-07-05T08:29:23.552886+00:00"}