{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:JEXUFH6MRIUPAGDENEG6APEYA3","short_pith_number":"pith:JEXUFH6M","schema_version":"1.0","canonical_sha256":"492f429fcc8a28f01864690de03c9806f67e90aaca8c77c93f40831f63996d66","source":{"kind":"arxiv","id":"2608.10260","version":1},"attestation_state":"computed","paper":{"title":"Interpreting Language Model Hidden States at Scale","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Daniel McKenzie, Ian Foster, Jordan Pettyjohn, Kyle Chard, Mansi Sakarvadia, Nathaniel Hudson","submitted_at":"2026-08-10T21:49:45Z","abstract_excerpt":"Lens methods interpret large language models (LLMs) by mapping intermediate activations to the output vocabulary, revealing how next-token predictions develop through the network. Trained lenses remain expensive: affine-translator parameters grow quadratically with model width, while exact, full-vocabulary Kullback--Leibler (KL) training dominates memory. Consequently, prior trained lenses have been applied to models of at most 20B parameters and remain tied to particular component types. We present OmniLens, which applies a single lens family to any model-width activation, whether residual st"},"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":"2608.10260","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-08-10T21:49:45Z","cross_cats_sorted":[],"title_canon_sha256":"11ccc42ea3bd49251e0537409a43f7d3e5e6de85200e2c4007174225790a5362","abstract_canon_sha256":"cbb2929aa2717e40a6587b5ee96870fb4951375ca98f9363351fe00a996cb644"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-12T00:23:12.281500Z","signature_b64":"Lz/hhmwqZiL+vAGEdYYjXptX4X5ZqsQVU4AJnuUTllXn9MYbIttQq9WKzSpXw2fJ0N56wZIe4De0e/Ik098gDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"492f429fcc8a28f01864690de03c9806f67e90aaca8c77c93f40831f63996d66","last_reissued_at":"2026-08-12T00:23:12.279954Z","signature_status":"signed_v1","first_computed_at":"2026-08-12T00:23:12.279954Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Interpreting Language Model Hidden States at Scale","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Daniel McKenzie, Ian Foster, Jordan Pettyjohn, Kyle Chard, Mansi Sakarvadia, Nathaniel Hudson","submitted_at":"2026-08-10T21:49:45Z","abstract_excerpt":"Lens methods interpret large language models (LLMs) by mapping intermediate activations to the output vocabulary, revealing how next-token predictions develop through the network. Trained lenses remain expensive: affine-translator parameters grow quadratically with model width, while exact, full-vocabulary Kullback--Leibler (KL) training dominates memory. Consequently, prior trained lenses have been applied to models of at most 20B parameters and remain tied to particular component types. We present OmniLens, which applies a single lens family to any model-width activation, whether residual st"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.10260","kind":"arxiv","version":1},"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/2608.10260/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":"2608.10260","created_at":"2026-08-12T00:23:12.282643+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.10260v1","created_at":"2026-08-12T00:23:12.282643+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.10260","created_at":"2026-08-12T00:23:12.282643+00:00"},{"alias_kind":"pith_short_12","alias_value":"JEXUFH6MRIUP","created_at":"2026-08-12T00:23:12.282643+00:00"},{"alias_kind":"pith_short_16","alias_value":"JEXUFH6MRIUPAGDE","created_at":"2026-08-12T00:23:12.282643+00:00"},{"alias_kind":"pith_short_8","alias_value":"JEXUFH6M","created_at":"2026-08-12T00:23:12.282643+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JEXUFH6MRIUPAGDENEG6APEYA3","json":"https://pith.science/pith/JEXUFH6MRIUPAGDENEG6APEYA3.json","graph_json":"https://pith.science/api/pith-number/JEXUFH6MRIUPAGDENEG6APEYA3/graph.json","events_json":"https://pith.science/api/pith-number/JEXUFH6MRIUPAGDENEG6APEYA3/events.json","paper":"https://pith.science/paper/JEXUFH6M"},"agent_actions":{"view_html":"https://pith.science/pith/JEXUFH6MRIUPAGDENEG6APEYA3","download_json":"https://pith.science/pith/JEXUFH6MRIUPAGDENEG6APEYA3.json","view_paper":"https://pith.science/paper/JEXUFH6M","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.10260&json=true","fetch_graph":"https://pith.science/api/pith-number/JEXUFH6MRIUPAGDENEG6APEYA3/graph.json","fetch_events":"https://pith.science/api/pith-number/JEXUFH6MRIUPAGDENEG6APEYA3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JEXUFH6MRIUPAGDENEG6APEYA3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JEXUFH6MRIUPAGDENEG6APEYA3/action/storage_attestation","attest_author":"https://pith.science/pith/JEXUFH6MRIUPAGDENEG6APEYA3/action/author_attestation","sign_citation":"https://pith.science/pith/JEXUFH6MRIUPAGDENEG6APEYA3/action/citation_signature","submit_replication":"https://pith.science/pith/JEXUFH6MRIUPAGDENEG6APEYA3/action/replication_record"}},"created_at":"2026-08-12T00:23:12.282643+00:00","updated_at":"2026-08-12T00:23:12.282643+00:00"}