{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:4KYGWJFFOKFGJ5WNIG5UH56I7Z","short_pith_number":"pith:4KYGWJFF","canonical_record":{"source":{"id":"2607.25132","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-07-27T23:00:53Z","cross_cats_sorted":[],"title_canon_sha256":"40ab6dcf28d4416a2514188d8000a7a91363f6888adddb59cdcd11a52fb44bd5","abstract_canon_sha256":"cc0e47251321db9f0d26d0dbfb785e87b0cffe380abfaa6df0d35ae09355862f"},"schema_version":"1.0"},"canonical_sha256":"e2b06b24a5728a64f6cd41bb43f7c8fe4ed001057596357964fe7de7e4d018db","source":{"kind":"arxiv","id":"2607.25132","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.25132","created_at":"2026-07-29T00:25:04Z"},{"alias_kind":"arxiv_version","alias_value":"2607.25132v1","created_at":"2026-07-29T00:25:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.25132","created_at":"2026-07-29T00:25:04Z"},{"alias_kind":"pith_short_12","alias_value":"4KYGWJFFOKFG","created_at":"2026-07-29T00:25:04Z"},{"alias_kind":"pith_short_16","alias_value":"4KYGWJFFOKFGJ5WN","created_at":"2026-07-29T00:25:04Z"},{"alias_kind":"pith_short_8","alias_value":"4KYGWJFF","created_at":"2026-07-29T00:25:04Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:4KYGWJFFOKFGJ5WNIG5UH56I7Z","target":"record","payload":{"canonical_record":{"source":{"id":"2607.25132","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-07-27T23:00:53Z","cross_cats_sorted":[],"title_canon_sha256":"40ab6dcf28d4416a2514188d8000a7a91363f6888adddb59cdcd11a52fb44bd5","abstract_canon_sha256":"cc0e47251321db9f0d26d0dbfb785e87b0cffe380abfaa6df0d35ae09355862f"},"schema_version":"1.0"},"canonical_sha256":"e2b06b24a5728a64f6cd41bb43f7c8fe4ed001057596357964fe7de7e4d018db","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-29T00:25:04.506791Z","signature_b64":"w3oB3H7Ei6RyKwzOoUPosBHCUIVcOEqCI7UzcI5b2whCWGPCJPYyMRRA0WGlM+9varxMidAXQgaGtnu4Tx4TBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e2b06b24a5728a64f6cd41bb43f7c8fe4ed001057596357964fe7de7e4d018db","last_reissued_at":"2026-07-29T00:25:04.505923Z","signature_status":"signed_v1","first_computed_at":"2026-07-29T00:25:04.505923Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.25132","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":"2026-07-29T00:25:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lnizPMmNJ0AxQNWAYKOh9x/Xy+a0dcQvjHD/rfpOwPCHodUWu8u5Ap7sP5b479ztTcTrgp2wqIxLWJRNQTW0Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T19:48:14.392456Z"},"content_sha256":"65756fb731c7c8547a4a766417b9417a802f59662104191135d2aca4fa60c023","schema_version":"1.0","event_id":"sha256:65756fb731c7c8547a4a766417b9417a802f59662104191135d2aca4fa60c023"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:4KYGWJFFOKFGJ5WNIG5UH56I7Z","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Interpretable GOHR Agents via Sparse Autoencoders","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Hao Wang, Jacob Feldman, Lazaros K. Gallos, Paul B. Kantor, Shiwei Tan, Vladimir Menkov, Weiyi Qin, Wentian Wang, Yusong Zhao","submitted_at":"2026-07-27T23:00:53Z","abstract_excerpt":"A central challenge in interpreting learned decision-making systems is to determine whether their internal representations contain concepts that help explain their behavior. We report interpretability experiments for a tokenized autoregressive Transformer agent in the Game of Hidden Rules (GOHR). We focus on a compact two-rule task in which both hidden rules map object shapes to target buckets, but with different permutations. The policy is trained on episodes sampled from these two hidden rules and then evaluated with fixed weights. It is never given a rule label and does not use an explicit "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.25132","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/2607.25132/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":"2026-07-29T00:25:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HfHUOewpsMEo+sR8a3oDAbTPqCcIOnHdNsFHfmLxrwDWqiPnTGwGpe8r4FjcEVCTlQLA/ea8oJZf5g6jEROTAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T19:48:14.393675Z"},"content_sha256":"e2fd0d9a6b821649ac6fd87d698cf4fd4ca2b2ffa1952015f3b2e12629773994","schema_version":"1.0","event_id":"sha256:e2fd0d9a6b821649ac6fd87d698cf4fd4ca2b2ffa1952015f3b2e12629773994"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4KYGWJFFOKFGJ5WNIG5UH56I7Z/bundle.json","state_url":"https://pith.science/pith/4KYGWJFFOKFGJ5WNIG5UH56I7Z/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4KYGWJFFOKFGJ5WNIG5UH56I7Z/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":"2026-08-08T19:48:14Z","links":{"resolver":"https://pith.science/pith/4KYGWJFFOKFGJ5WNIG5UH56I7Z","bundle":"https://pith.science/pith/4KYGWJFFOKFGJ5WNIG5UH56I7Z/bundle.json","state":"https://pith.science/pith/4KYGWJFFOKFGJ5WNIG5UH56I7Z/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4KYGWJFFOKFGJ5WNIG5UH56I7Z/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:4KYGWJFFOKFGJ5WNIG5UH56I7Z","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":"cc0e47251321db9f0d26d0dbfb785e87b0cffe380abfaa6df0d35ae09355862f","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-07-27T23:00:53Z","title_canon_sha256":"40ab6dcf28d4416a2514188d8000a7a91363f6888adddb59cdcd11a52fb44bd5"},"schema_version":"1.0","source":{"id":"2607.25132","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.25132","created_at":"2026-07-29T00:25:04Z"},{"alias_kind":"arxiv_version","alias_value":"2607.25132v1","created_at":"2026-07-29T00:25:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.25132","created_at":"2026-07-29T00:25:04Z"},{"alias_kind":"pith_short_12","alias_value":"4KYGWJFFOKFG","created_at":"2026-07-29T00:25:04Z"},{"alias_kind":"pith_short_16","alias_value":"4KYGWJFFOKFGJ5WN","created_at":"2026-07-29T00:25:04Z"},{"alias_kind":"pith_short_8","alias_value":"4KYGWJFF","created_at":"2026-07-29T00:25:04Z"}],"graph_snapshots":[{"event_id":"sha256:e2fd0d9a6b821649ac6fd87d698cf4fd4ca2b2ffa1952015f3b2e12629773994","target":"graph","created_at":"2026-07-29T00:25:04Z","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/2607.25132/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"A central challenge in interpreting learned decision-making systems is to determine whether their internal representations contain concepts that help explain their behavior. We report interpretability experiments for a tokenized autoregressive Transformer agent in the Game of Hidden Rules (GOHR). We focus on a compact two-rule task in which both hidden rules map object shapes to target buckets, but with different permutations. The policy is trained on episodes sampled from these two hidden rules and then evaluated with fixed weights. It is never given a rule label and does not use an explicit ","authors_text":"Hao Wang, Jacob Feldman, Lazaros K. Gallos, Paul B. Kantor, Shiwei Tan, Vladimir Menkov, Weiyi Qin, Wentian Wang, Yusong Zhao","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-07-27T23:00:53Z","title":"Interpretable GOHR Agents via Sparse Autoencoders"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.25132","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":"sha256:65756fb731c7c8547a4a766417b9417a802f59662104191135d2aca4fa60c023","target":"record","created_at":"2026-07-29T00:25:04Z","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":"cc0e47251321db9f0d26d0dbfb785e87b0cffe380abfaa6df0d35ae09355862f","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-07-27T23:00:53Z","title_canon_sha256":"40ab6dcf28d4416a2514188d8000a7a91363f6888adddb59cdcd11a52fb44bd5"},"schema_version":"1.0","source":{"id":"2607.25132","kind":"arxiv","version":1}},"canonical_sha256":"e2b06b24a5728a64f6cd41bb43f7c8fe4ed001057596357964fe7de7e4d018db","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e2b06b24a5728a64f6cd41bb43f7c8fe4ed001057596357964fe7de7e4d018db","first_computed_at":"2026-07-29T00:25:04.505923Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-29T00:25:04.505923Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"w3oB3H7Ei6RyKwzOoUPosBHCUIVcOEqCI7UzcI5b2whCWGPCJPYyMRRA0WGlM+9varxMidAXQgaGtnu4Tx4TBQ==","signature_status":"signed_v1","signed_at":"2026-07-29T00:25:04.506791Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.25132","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:65756fb731c7c8547a4a766417b9417a802f59662104191135d2aca4fa60c023","sha256:e2fd0d9a6b821649ac6fd87d698cf4fd4ca2b2ffa1952015f3b2e12629773994"],"state_sha256":"90abade6e543164357e6723e700f2e277278b244ee8ab9db9f0a1bb7b1757ac8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ifEH+XA9QB46srI+yggECGD+Wr6N1g70M7a0kfeTm24aiT+NLZPqSfWkpb5bX7bxcna73qSkw1o4sSEpmnnrDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T19:48:14.399697Z","bundle_sha256":"5f5ddbfe40a6acee213bb3a7350c84bd66001def48389e29199f2b61faed69c8"}}