{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:4AGCKVPO4ZZECCAETM5CL4NMLL","short_pith_number":"pith:4AGCKVPO","canonical_record":{"source":{"id":"2312.01037","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-12-02T05:47:22Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"ef924b98fa61e63f8754fc0fb2b38d4eda866019b9b326f96c48355f83a2342e","abstract_canon_sha256":"eb8bb47e4027f4c470844dfb80ad53422d42a8fc20114a975d86593318653adb"},"schema_version":"1.0"},"canonical_sha256":"e00c2555eee6724108049b3a25f1ac5aed4f6e2d57089c1f2183192d455c26fa","source":{"kind":"arxiv","id":"2312.01037","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2312.01037","created_at":"2026-07-05T08:53:52Z"},{"alias_kind":"arxiv_version","alias_value":"2312.01037v4","created_at":"2026-07-05T08:53:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.01037","created_at":"2026-07-05T08:53:52Z"},{"alias_kind":"pith_short_12","alias_value":"4AGCKVPO4ZZE","created_at":"2026-07-05T08:53:52Z"},{"alias_kind":"pith_short_16","alias_value":"4AGCKVPO4ZZECCAE","created_at":"2026-07-05T08:53:52Z"},{"alias_kind":"pith_short_8","alias_value":"4AGCKVPO","created_at":"2026-07-05T08:53:52Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:4AGCKVPO4ZZECCAETM5CL4NMLL","target":"record","payload":{"canonical_record":{"source":{"id":"2312.01037","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-12-02T05:47:22Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"ef924b98fa61e63f8754fc0fb2b38d4eda866019b9b326f96c48355f83a2342e","abstract_canon_sha256":"eb8bb47e4027f4c470844dfb80ad53422d42a8fc20114a975d86593318653adb"},"schema_version":"1.0"},"canonical_sha256":"e00c2555eee6724108049b3a25f1ac5aed4f6e2d57089c1f2183192d455c26fa","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:53:52.568717Z","signature_b64":"7xRDeb/6B3h8WetwU7XagtL7JTTZ0tcJoPsDS3MUCmFbkCZXHX5qJbkwjB0FHRc1pWJ6R65VTm7BCQ2qXHWvBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e00c2555eee6724108049b3a25f1ac5aed4f6e2d57089c1f2183192d455c26fa","last_reissued_at":"2026-07-05T08:53:52.568262Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:53:52.568262Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2312.01037","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:53:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"h88V3gkLy94UBriaHsbFXUT+QILb/+4sy3/x00L1O87GHELGMKAYy8UoVQTh09pHWxQR4VXQdF4wVdEOmIT0Aw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T13:22:48.996071Z"},"content_sha256":"9491248a61d30257f3a554996c482c7327eec935efa89ae9d965c526432cac85","schema_version":"1.0","event_id":"sha256:9491248a61d30257f3a554996c482c7327eec935efa89ae9d965c526432cac85"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:4AGCKVPO4ZZECCAETM5CL4NMLL","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Eliciting Latent Knowledge from Quirky Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Alex Mallen, Julia Kharchenko, Madeline Brumley, Nora Belrose","submitted_at":"2023-12-02T05:47:22Z","abstract_excerpt":"Eliciting Latent Knowledge (ELK) aims to find patterns in a capable neural network's activations that robustly track the true state of the world, especially in hard-to-verify cases where the model's output is untrusted. To further ELK research, we introduce 12 datasets and a corresponding suite of \"quirky\" language models (LMs) that are finetuned to make systematic errors when answering questions if and only if the keyword \"Bob\" is present in the prompt. We find that, especially in middle layers, linear probes usually report an LM's knowledge independently of what the LM outputs, enabling us t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.01037","kind":"arxiv","version":4},"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/2312.01037/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":"2026-07-05T08:53:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5Z7O60bxzGezUEEdItSxmLPSX7XrZ46OIrfahUC9kcWIobgyIGkjzZu+Pn0bXHKRtX3+kCT5aQpFRQw/FpraAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T13:22:48.996556Z"},"content_sha256":"ec5174eae5946a8cf5d8e079345b4156825fce01f763a8f8f8ad8df508f5568c","schema_version":"1.0","event_id":"sha256:ec5174eae5946a8cf5d8e079345b4156825fce01f763a8f8f8ad8df508f5568c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4AGCKVPO4ZZECCAETM5CL4NMLL/bundle.json","state_url":"https://pith.science/pith/4AGCKVPO4ZZECCAETM5CL4NMLL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4AGCKVPO4ZZECCAETM5CL4NMLL/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":"2026-08-15T13:22:49Z","links":{"resolver":"https://pith.science/pith/4AGCKVPO4ZZECCAETM5CL4NMLL","bundle":"https://pith.science/pith/4AGCKVPO4ZZECCAETM5CL4NMLL/bundle.json","state":"https://pith.science/pith/4AGCKVPO4ZZECCAETM5CL4NMLL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4AGCKVPO4ZZECCAETM5CL4NMLL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:4AGCKVPO4ZZECCAETM5CL4NMLL","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":"eb8bb47e4027f4c470844dfb80ad53422d42a8fc20114a975d86593318653adb","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-12-02T05:47:22Z","title_canon_sha256":"ef924b98fa61e63f8754fc0fb2b38d4eda866019b9b326f96c48355f83a2342e"},"schema_version":"1.0","source":{"id":"2312.01037","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2312.01037","created_at":"2026-07-05T08:53:52Z"},{"alias_kind":"arxiv_version","alias_value":"2312.01037v4","created_at":"2026-07-05T08:53:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.01037","created_at":"2026-07-05T08:53:52Z"},{"alias_kind":"pith_short_12","alias_value":"4AGCKVPO4ZZE","created_at":"2026-07-05T08:53:52Z"},{"alias_kind":"pith_short_16","alias_value":"4AGCKVPO4ZZECCAE","created_at":"2026-07-05T08:53:52Z"},{"alias_kind":"pith_short_8","alias_value":"4AGCKVPO","created_at":"2026-07-05T08:53:52Z"}],"graph_snapshots":[{"event_id":"sha256:ec5174eae5946a8cf5d8e079345b4156825fce01f763a8f8f8ad8df508f5568c","target":"graph","created_at":"2026-07-05T08:53:52Z","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/2312.01037/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Eliciting Latent Knowledge (ELK) aims to find patterns in a capable neural network's activations that robustly track the true state of the world, especially in hard-to-verify cases where the model's output is untrusted. To further ELK research, we introduce 12 datasets and a corresponding suite of \"quirky\" language models (LMs) that are finetuned to make systematic errors when answering questions if and only if the keyword \"Bob\" is present in the prompt. We find that, especially in middle layers, linear probes usually report an LM's knowledge independently of what the LM outputs, enabling us t","authors_text":"Alex Mallen, Julia Kharchenko, Madeline Brumley, Nora Belrose","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-12-02T05:47:22Z","title":"Eliciting Latent Knowledge from Quirky Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.01037","kind":"arxiv","version":4},"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":"sha256:9491248a61d30257f3a554996c482c7327eec935efa89ae9d965c526432cac85","target":"record","created_at":"2026-07-05T08:53:52Z","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":"eb8bb47e4027f4c470844dfb80ad53422d42a8fc20114a975d86593318653adb","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-12-02T05:47:22Z","title_canon_sha256":"ef924b98fa61e63f8754fc0fb2b38d4eda866019b9b326f96c48355f83a2342e"},"schema_version":"1.0","source":{"id":"2312.01037","kind":"arxiv","version":4}},"canonical_sha256":"e00c2555eee6724108049b3a25f1ac5aed4f6e2d57089c1f2183192d455c26fa","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e00c2555eee6724108049b3a25f1ac5aed4f6e2d57089c1f2183192d455c26fa","first_computed_at":"2026-07-05T08:53:52.568262Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:53:52.568262Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7xRDeb/6B3h8WetwU7XagtL7JTTZ0tcJoPsDS3MUCmFbkCZXHX5qJbkwjB0FHRc1pWJ6R65VTm7BCQ2qXHWvBA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:53:52.568717Z","signed_message":"canonical_sha256_bytes"},"source_id":"2312.01037","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9491248a61d30257f3a554996c482c7327eec935efa89ae9d965c526432cac85","sha256:ec5174eae5946a8cf5d8e079345b4156825fce01f763a8f8f8ad8df508f5568c"],"state_sha256":"9fa88bf6d03f832014d7486edc9cf7ed1903d2922599f293dec46a9e90216ae8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"EV04C+jyiushLW6vgXdHFypL3K5fBFwy4yx86EkkOk5L7Zxgw7zdeIN6qYTfgEt8tkczkPhL2lB0Dwi/HDgcDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-15T13:22:49.002056Z","bundle_sha256":"f96072c1659521615d227adb28939294536277733db559174847a0d46567ad4e"}}