{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:YXMTQON5DLYJZKZJHSLMCYCYCN","short_pith_number":"pith:YXMTQON5","canonical_record":{"source":{"id":"2502.09135","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-13T10:11:36Z","cross_cats_sorted":["q-bio.BM"],"title_canon_sha256":"87872298731358d3c0e400a88f6b06aa4dfe8138aca0a395fc1810cc1eb355df","abstract_canon_sha256":"db78decc33c5f65e2553170c79abfe62e44d6a0c1c2806d7e60051a961901886"},"schema_version":"1.0"},"canonical_sha256":"c5d93839bd1af09cab293c96c160581347c576f7c305b7110bc0e70b45a82feb","source":{"kind":"arxiv","id":"2502.09135","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.09135","created_at":"2026-07-05T10:13:52Z"},{"alias_kind":"arxiv_version","alias_value":"2502.09135v1","created_at":"2026-07-05T10:13:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.09135","created_at":"2026-07-05T10:13:52Z"},{"alias_kind":"pith_short_12","alias_value":"YXMTQON5DLYJ","created_at":"2026-07-05T10:13:52Z"},{"alias_kind":"pith_short_16","alias_value":"YXMTQON5DLYJZKZJ","created_at":"2026-07-05T10:13:52Z"},{"alias_kind":"pith_short_8","alias_value":"YXMTQON5","created_at":"2026-07-05T10:13:52Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:YXMTQON5DLYJZKZJHSLMCYCYCN","target":"record","payload":{"canonical_record":{"source":{"id":"2502.09135","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-13T10:11:36Z","cross_cats_sorted":["q-bio.BM"],"title_canon_sha256":"87872298731358d3c0e400a88f6b06aa4dfe8138aca0a395fc1810cc1eb355df","abstract_canon_sha256":"db78decc33c5f65e2553170c79abfe62e44d6a0c1c2806d7e60051a961901886"},"schema_version":"1.0"},"canonical_sha256":"c5d93839bd1af09cab293c96c160581347c576f7c305b7110bc0e70b45a82feb","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:13:52.843464Z","signature_b64":"WdBYFrQ/Hvq/HHTY4p1W39qoh1vv1ynCe7RPsqRx4ebRyXNBDRJAGbKL19JEDT+l92CpwB4g4xNTWD1v/aQ6Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c5d93839bd1af09cab293c96c160581347c576f7c305b7110bc0e70b45a82feb","last_reissued_at":"2026-07-05T10:13:52.842969Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:13:52.842969Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.09135","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:13:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wrP7en8I0+aZM2FtmJHYOVRCZ3R+AC2R0HlJD/03kxR1C5XJry1UfHZrgzOJGPUsYAGyRETMvQfgXvfo5STAAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T13:59:03.222770Z"},"content_sha256":"757bdf00a8e2349ec39f52b4477fb3426eac4b7cea5b3ffaa6ef7520d1d1a4e4","schema_version":"1.0","event_id":"sha256:757bdf00a8e2349ec39f52b4477fb3426eac4b7cea5b3ffaa6ef7520d1d1a4e4"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:YXMTQON5DLYJZKZJHSLMCYCYCN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Interpreting and Steering Protein Language Models through Sparse Autoencoders","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["q-bio.BM"],"primary_cat":"cs.LG","authors_text":"Alessio Ansuini, Edith Natalia Villegas Garcia","submitted_at":"2025-02-13T10:11:36Z","abstract_excerpt":"The rapid advancements in transformer-based language models have revolutionized natural language processing, yet understanding the internal mechanisms of these models remains a significant challenge. This paper explores the application of sparse autoencoders (SAE) to interpret the internal representations of protein language models, specifically focusing on the ESM-2 8M parameter model. By performing a statistical analysis on each latent component's relevance to distinct protein annotations, we identify potential interpretations linked to various protein characteristics, including transmembran"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.09135","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/2502.09135/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":"2026-07-05T10:13:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9pWdky/2Rc/8H5z4wLvEbAzpRIj8Ee8ljh7SJnzPCoHVO403wyOCfFwCKA1ctA0VesinMTFKgLaRwlDOkTifBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T13:59:03.223416Z"},"content_sha256":"046cc1d27b86c6b07195c12aa23ba067d3d2eb140ece78aebfb98feae9d84754","schema_version":"1.0","event_id":"sha256:046cc1d27b86c6b07195c12aa23ba067d3d2eb140ece78aebfb98feae9d84754"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YXMTQON5DLYJZKZJHSLMCYCYCN/bundle.json","state_url":"https://pith.science/pith/YXMTQON5DLYJZKZJHSLMCYCYCN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YXMTQON5DLYJZKZJHSLMCYCYCN/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":"2026-08-20T13:59:03Z","links":{"resolver":"https://pith.science/pith/YXMTQON5DLYJZKZJHSLMCYCYCN","bundle":"https://pith.science/pith/YXMTQON5DLYJZKZJHSLMCYCYCN/bundle.json","state":"https://pith.science/pith/YXMTQON5DLYJZKZJHSLMCYCYCN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YXMTQON5DLYJZKZJHSLMCYCYCN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:YXMTQON5DLYJZKZJHSLMCYCYCN","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":"db78decc33c5f65e2553170c79abfe62e44d6a0c1c2806d7e60051a961901886","cross_cats_sorted":["q-bio.BM"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-13T10:11:36Z","title_canon_sha256":"87872298731358d3c0e400a88f6b06aa4dfe8138aca0a395fc1810cc1eb355df"},"schema_version":"1.0","source":{"id":"2502.09135","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.09135","created_at":"2026-07-05T10:13:52Z"},{"alias_kind":"arxiv_version","alias_value":"2502.09135v1","created_at":"2026-07-05T10:13:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.09135","created_at":"2026-07-05T10:13:52Z"},{"alias_kind":"pith_short_12","alias_value":"YXMTQON5DLYJ","created_at":"2026-07-05T10:13:52Z"},{"alias_kind":"pith_short_16","alias_value":"YXMTQON5DLYJZKZJ","created_at":"2026-07-05T10:13:52Z"},{"alias_kind":"pith_short_8","alias_value":"YXMTQON5","created_at":"2026-07-05T10:13:52Z"}],"graph_snapshots":[{"event_id":"sha256:046cc1d27b86c6b07195c12aa23ba067d3d2eb140ece78aebfb98feae9d84754","target":"graph","created_at":"2026-07-05T10:13:52Z","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/2502.09135/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The rapid advancements in transformer-based language models have revolutionized natural language processing, yet understanding the internal mechanisms of these models remains a significant challenge. This paper explores the application of sparse autoencoders (SAE) to interpret the internal representations of protein language models, specifically focusing on the ESM-2 8M parameter model. By performing a statistical analysis on each latent component's relevance to distinct protein annotations, we identify potential interpretations linked to various protein characteristics, including transmembran","authors_text":"Alessio Ansuini, Edith Natalia Villegas Garcia","cross_cats":["q-bio.BM"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-13T10:11:36Z","title":"Interpreting and Steering Protein Language Models through Sparse Autoencoders"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.09135","kind":"arxiv","version":1},"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":"sha256:757bdf00a8e2349ec39f52b4477fb3426eac4b7cea5b3ffaa6ef7520d1d1a4e4","target":"record","created_at":"2026-07-05T10:13:52Z","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":"db78decc33c5f65e2553170c79abfe62e44d6a0c1c2806d7e60051a961901886","cross_cats_sorted":["q-bio.BM"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-13T10:11:36Z","title_canon_sha256":"87872298731358d3c0e400a88f6b06aa4dfe8138aca0a395fc1810cc1eb355df"},"schema_version":"1.0","source":{"id":"2502.09135","kind":"arxiv","version":1}},"canonical_sha256":"c5d93839bd1af09cab293c96c160581347c576f7c305b7110bc0e70b45a82feb","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c5d93839bd1af09cab293c96c160581347c576f7c305b7110bc0e70b45a82feb","first_computed_at":"2026-07-05T10:13:52.842969Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:13:52.842969Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"WdBYFrQ/Hvq/HHTY4p1W39qoh1vv1ynCe7RPsqRx4ebRyXNBDRJAGbKL19JEDT+l92CpwB4g4xNTWD1v/aQ6Dw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:13:52.843464Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.09135","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:757bdf00a8e2349ec39f52b4477fb3426eac4b7cea5b3ffaa6ef7520d1d1a4e4","sha256:046cc1d27b86c6b07195c12aa23ba067d3d2eb140ece78aebfb98feae9d84754"],"state_sha256":"982562dc449ae107990b37f83b348bef2f1f77c22f7a728f631e3cacc4862bdd"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MgHEZItMGij/TON3JKeFXb/xhl+I4gC0VZCL0/gdFH522irQkrmL+7V9JURQWZIIqWYeLkvwJqnXQOYPD1NnDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-20T13:59:03.228756Z","bundle_sha256":"a0bd837e59d8ff2145f13b04e688c2e928a80982710117aa2fcde15e9287b176"}}