{"as_of":"2026-08-17T20:42:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:51a53a61e6c80ffc68977e0a9f3103b498b4e94fe23d33ea054a2573fa9e40df","coverage":[{"denominator":24,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":24,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-16T12:29:34.141936Z","state":"measured"},{"denominator":24,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":24,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-17T06:30:58.91139+00:00","state":"measured"},{"denominator":0,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":0,"source":"paper_references, paper_reference_links","source_observed_at":null,"state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"cited_works","source_observed_at":null,"state":"measured"}],"external_citation_measurements":[],"inbound":[],"links":{"evidence":"/evidence","html":"/paper/2504.12939/citation-record","integrity":"/paper/2504.12939/integrity","json":"/paper/2504.12939/citation-record.json","paper":"/paper/2504.12939"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:29:34.658849Z","title":"On pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation","venue":null,"work_id":"6d49443e-5f71-4a89-a57a-a6e5b46d6eec","year":null},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.006697Z"},"links":{"citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:62b46b59e9ebcc8b72d4a138b4de19240d04c49fd6e10d0285aa3e14f6cb68b0","observation_id":"9bc9122c-2d72-452a-91d0-37ab95944ede","resolution":{"observed_at":"2026-08-16T12:29:34.664511Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:29:34.641248Z","title":"Concept whitening for interpretable image recognition","venue":null,"work_id":"11ada3bd-71eb-4d30-ac3d-9857d5d0ab7b","year":2020},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.012779Z"},"links":{"citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:f89a5d3b8590a579fdf2f8ccc81f863ab2da94f83224c0de1aecb4f889a0fe4f","observation_id":"685e1138-e4d4-409c-b31f-32f9ae98b4ec","resolution":{"observed_at":"2026-08-16T12:29:34.646742Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:29:34.623645Z","title":"ImageNet: A large-scale hierarchical image database","venue":null,"work_id":"dd89de83-bffa-4b78-92d6-bc4497a2514f","year":2009},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.018372Z"},"links":{"citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:bec664341e838f9ae3103cb6fd9cd2b8a43a976ff524bfa1a4b757ae33f48965","observation_id":"70d894a7-61fe-4e14-a8de-35f7bdd2c906","resolution":{"observed_at":"2026-08-16T12:29:34.629228Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:29:34.606427Z","title":"Visual and semantic similarity in ImageNet","venue":null,"work_id":"fbad63da-60db-40e4-b75c-3863439815b0","year":2011},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.023775Z"},"links":{"citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:510c9300731b5ab07f28ba2201c568a14d67758f4081bddfc109a0dfd5f9ccc3","observation_id":"28ceea0a-2800-4040-82af-5fa3eecfde19","resolution":{"observed_at":"2026-08-16T12:29:34.611958Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:29:34.588960Z","title":"PURE: Turning polysemantic neurons into pure features by identifying rele- vant circuits","venue":null,"work_id":"3d064801-7aaa-40d1-b2b7-723989909715","year":null},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.029553Z"},"links":{"citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:129d674a9c623b7be85e7c5c0a1539285ca24be894f550775bc5804406f41c4d","observation_id":"4065f58c-07ed-4119-aac8-016b992e109c","resolution":{"observed_at":"2026-08-16T12:29:34.594765Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:29:34.569259Z","title":"Toy models of superposition","venue":null,"work_id":"526a4650-514a-4934-9318-506e52893b68","year":2022},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.035654Z"},"links":{"citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:29944c5f2834a536fa681a7fceaed125b908adf15ec319b025f57e3d1afcbd85","observation_id":"31531fe7-ba9b-4948-a15a-b905382e19ba","resolution":{"observed_at":"2026-08-16T12:29:34.575950Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:29:34.549864Z","title":"Bengio, Aaron Courville, and Pascal Vin- cent","venue":null,"work_id":"83e5577e-0236-4ae3-96c2-ca12ae64c807","year":2009},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.041368Z"},"links":{"citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:0ae297a5c74c41c966d2122970184acc3fb65a00e4f52f022a0b6a52fcb18f36","observation_id":"68a34062-c96f-482e-9624-3e2e684528bd","resolution":{"observed_at":"2026-08-16T12:29:34.555654Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:29:34.532512Z","title":"Unlocking feature visualization for deep net- work with magnitude constrained optimization","venue":null,"work_id":"391f025c-f899-4c81-9aa5-ce3b0c74d1c2","year":2023},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.046772Z"},"links":{"citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:37ccaab02048f854519766c1fcd4562800d0a06d787a77bd51846f10a687e844","observation_id":"f9532686-f7c0-40f9-8c4a-c24969bd5c87","resolution":{"observed_at":"2026-08-16T12:29:34.538119Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:29:34.514252Z","title":"Deep residual learning for image recognition","venue":null,"work_id":"4a488a84-6b03-4b12-9d40-4e1b90036d3e","year":2016},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.053879Z"},"links":{"citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:b046460d188c4b932897f0c21085cf77b4a36216f03274a469201175f828614c","observation_id":"07ccb376-c936-4588-a4bc-8413a996caa0","resolution":{"observed_at":"2026-08-16T12:29:34.520041Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:29:34.059714Z","title":"Sparse autoencoders can interpret randomly ini- tialized transformers","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.059714Z"},"links":{"citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:1695c0798e1638fe5a3a499717193a7aa46bdbe854154013c19d1cc8e5899ccf","observation_id":"51a98665-09a0-451c-8169-6212020005cc","resolution":{"observed_at":"2026-08-16T12:29:34.059714Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:29:34.496739Z","title":"Sparse autoencoders find highly interpretable features in language models","venue":null,"work_id":"7dbe6b80-1c72-46ca-addd-5acf766043a4","year":2024},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.065375Z"},"links":{"citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:9bc02b2f3e7e6b033ab584d96dab67cc03f1d6fa35bf89ed747e785048796904","observation_id":"e55e1aa4-8b84-4b79-8d2b-c74a41cb8917","resolution":{"observed_at":"2026-08-16T12:29:34.502036Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:29:34.478729Z","title":"Cai, James Wexler, Fernanda B","venue":null,"work_id":"4df47059-f9b9-4e61-9ae5-30635cb160e5","year":2018},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.071714Z"},"links":{"citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:d1e1f500482ad2e4190e2c5a907a1402c6d5d0d1042204df0366fd6390cd8b62","observation_id":"b1df342a-90dd-41fb-8085-949328e082b7","resolution":{"observed_at":"2026-08-16T12:29:34.484321Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:29:34.077304Z","title":null,"venue":null,"work_id":null,"year":1992},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.077304Z"},"links":{"citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:9ef4ee868b20af5a2c2d4358ee922a1dcc183591e8a73a2e4558f7a5fe3f5aa9","observation_id":"ed25ec22-3ad1-4f43-b99f-4a0492856683","resolution":{"observed_at":"2026-08-16T12:29:34.077304Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:29:34.450186Z","title":"Compositional explanations of neurons","venue":null,"work_id":"f9223df0-e72e-4473-89b1-0d29a3cfc2a5","year":2020},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.083020Z"},"links":{"citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:caec2d5be6e2e89d4fd3b0897730ec8c59eeaab9d46438cb7157da8475c6584a","observation_id":"b0e8f753-da50-485a-833e-4bd29552a10a","resolution":{"observed_at":"2026-08-16T12:29:34.455359Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1602.03616","last_updated":"2016-05-07T06:30:51Z","snapshot_observed_at":"2026-08-14T22:10:48.137147Z","submitted_at":"2016-02-11T05:10:42Z","title":"Multifaceted Feature Visualization: Uncovering the Different Types of Features Learned By Each Neuron in Deep Neural Networks","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1602.03616","snapshot_observed_at":"2026-08-16T12:29:34.088764Z","title":"Multi- faceted feature visualization: Uncovering the different types of features learned by each neuron in deep neural networks","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.088764Z"},"links":{"cited_paper":"/paper/1602.03616","citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:e9225e1aa070ceb00090931a9bb922842f0a576b858f160720ac2ac037dfc264","observation_id":"399c0446-61ab-4605-ae3a-d6d665092e33","resolution":{"observed_at":"2026-08-16T12:29:34.088764Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2405.06855","last_updated":"2024-05-10T23:48:37Z","snapshot_observed_at":"2026-08-16T13:53:36.627299Z","submitted_at":"2024-05-10T23:48:37Z","title":"Linear Explanations for Individual Neurons","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2405.06855","snapshot_observed_at":"2026-08-16T12:29:34.094557Z","title":"Oikarinen and Tsui-Wei Weng","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.094557Z"},"links":{"cited_paper":"/paper/2405.06855","citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:e03d533466b17fd4647e4406f01b5e11967a1f3506ee35d681e13bf76b045660","observation_id":"9f1f9bbd-834e-4bca-9bf0-fdd36f5eb5c5","resolution":{"observed_at":"2026-08-16T12:29:34.094557Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:29:34.430986Z","title":"Feature visualization","venue":null,"work_id":"6a79be60-f377-4fbd-912f-b5f03ee0c083","year":2017},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.100630Z"},"links":{"citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:be7ae61695c584da4b51fdf18e149d1fc1014735b557a1d8293c966fe41ddc20","observation_id":"0f10d64b-abe1-4f8d-b227-8a906385a2fa","resolution":{"observed_at":"2026-08-16T12:29:34.436933Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:29:34.410637Z","title":"Disentangling neuron representations with concept vectors","venue":null,"work_id":"60bba17f-8ae4-4c69-8d04-ac49e1122fa4","year":2023},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.106882Z"},"links":{"citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:ae7793ba74ec4e229f951e37e86088baed272ce6556496233689c588115cea27","observation_id":"97c48b91-ef1c-4e20-bb1f-e7daa10fa90d","resolution":{"observed_at":"2026-08-16T12:29:34.416112Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:29:34.391430Z","title":"Automatic differentiation in PyTorch","venue":null,"work_id":"0fc57c32-c530-44e8-9ca9-09498999e1e6","year":2017},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.112126Z"},"links":{"citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:3a6a3392084df7bdf596a45f58e656290155639230bf5e47d40687515b964235","observation_id":"f2760fae-ac7b-42e6-a916-9f0a3f502597","resolution":{"observed_at":"2026-08-16T12:29:34.396747Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2501.16615","last_updated":"2025-01-29T19:18:45Z","snapshot_observed_at":"2026-08-15T16:41:45.621359Z","submitted_at":"2025-01-28T01:24:16Z","title":"Sparse Autoencoders Trained on the Same Data Learn Different Features","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2501.16615","snapshot_observed_at":"2026-08-16T12:29:34.117929Z","title":"Sparse autoencoders trained on the same data learn different features","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.117929Z"},"links":{"cited_paper":"/paper/2501.16615","citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:1da05d887d3083f297f5a12a5280098cb85bb516e7272c5c4155e1e3ec32dbbd","observation_id":"5a507f8f-967b-44f5-bff3-5a3cd4c269fa","resolution":{"observed_at":"2026-08-16T12:29:34.117929Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:29:34.372879Z","title":"Towards a fuller understanding of neurons with clustered compositional explanations","venue":null,"work_id":"2587cfc6-a541-4ed2-8f01-a15ae2193f30","year":null},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.123572Z"},"links":{"citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:5e59fe9cd9d3e9e9f3c1ebdcce1194f747c4d0cf1880ce49e3f5e354686c22a9","observation_id":"745bae8c-3e29-49e9-8d20-8e90aadab802","resolution":{"observed_at":"2026-08-16T12:29:34.378699Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:29:34.353006Z","title":"Learning important features through propagating activation differences","venue":null,"work_id":"1bc5ead9-427a-4bea-9883-f21f4345a4db","year":2017},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.129701Z"},"links":{"citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:ea1b32d2c09690f7ea1cf575a6a1b8a5edb83e1d12d96ecf1f48d657856b53cb","observation_id":"68bc6bdf-e296-4d65-9f49-44d57fb7a6b5","resolution":{"observed_at":"2026-08-16T12:29:34.359952Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:29:34.329785Z","title":"Axiomatic attribution for deep networks","venue":null,"work_id":"4bd37b47-b1a0-4cf5-992c-3a3266262644","year":null},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.135827Z"},"links":{"citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:b0fe303710cdc3192bafc9646e56237aad81fe68f02104b388c04d5b3b45fd07","observation_id":"2a706954-7cef-4d2a-b50a-788376a46795","resolution":{"observed_at":"2026-08-16T12:29:34.337586Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2501.17148","last_updated":"2025-03-03T21:15:30Z","snapshot_observed_at":"2026-08-16T12:58:32.947185Z","submitted_at":"2025-01-28T18:51:24Z","title":"AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2501.17148","snapshot_observed_at":"2026-08-16T12:29:34.141936Z","title":"Man- ning, and Christopher Potts","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-16T12:29:34.141936Z"},"links":{"cited_paper":"/paper/2501.17148","citing_paper":"/paper/2504.12939"},"observation_digest":"sha256:a5662095f15b0959e2f35c427b74b8638bff7f2a335c818e0704dd07295c8392","observation_id":"11c5b09b-adb1-4bb7-8871-25e75ff5c78f","resolution":{"observed_at":"2026-08-16T12:29:34.141936Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2504.12939","last_updated":"2025-04-17T13:37:47Z","latest_version":1,"primary_category":"cs.CV","snapshot_observed_at":"2026-08-16T12:16:51.698354Z","submitted_at":"2025-04-17T13:37:47Z","title":"Disentangling Polysemantic Channels in Convolutional Neural Networks"},"reference_resolution":{"displayed":24,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":6,"verified_exact":0,"verified_fuzzy":18},"total_outbound_references":24},"refusal":"A citation records a reference. It does not transfer a finding from one paper to another.","schema":"pith.paper-citation-record.v1","standing_sources":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"thesis":"As of 17 August 2026, this Paper Citation Record lists 24 of 24 outbound references and 0 inbound Pith citation observations for arXiv:2504.12939."}