{"as_of":"2026-08-10T00:33:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:595caa537c905a49df82728aee0006d96abb2f34dcb4e2e0898cf5fa817b58f4","coverage":[{"denominator":45,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":45,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-07T05:19:30.202178Z","state":"measured"},{"denominator":45,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":45,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-09T06:31:02.800959+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/2506.08359/citation-record","integrity":"/paper/2506.08359/integrity","json":"/paper/2506.08359/citation-record.json","paper":"/paper/2506.08359"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-07T05:19:30.638082Z","title":"Viégas, Hanspeter Pfister, and Martin Wattenberg","venue":null,"work_id":"ccf87a7e-bef7-4c82-b5d4-28cc371822bc","year":2023},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.042274Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:2dfca6a73523f2742124f82246f25aef5c1bc8fe0e8d97dbc25314f1af99bd9b","observation_id":"682dde86-4774-459c-8dae-7418cdaaff18","resolution":{"observed_at":"2026-08-07T05:19:30.641408Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+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":"2026-08-07T05:19:30.628669Z","title":"Steering gpt-2-xl by adding an activation vector","venue":null,"work_id":"7e9f13fc-65cd-40cd-8a18-d77d6d30397b","year":2023},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.047029Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:2d8fb36433412d2e742fab202757d94fb8124c3754b99f6bbc31f6ee9a8a2378","observation_id":"6d7153da-7d78-4441-adcf-7a07e2b50523","resolution":{"observed_at":"2026-08-07T05:19:30.632036Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+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":"2026-08-07T05:19:30.618427Z","title":"Steering llama 2 via contrastive activation addition","venue":null,"work_id":"4aa675c4-c1ea-4b77-be15-3f3b233a27ba","year":2024},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.050459Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:473ce9f0bcc005cce520a6670c0d306e5500a97ca125261dfc4e5e4aa6b74ebe","observation_id":"9cd0c6e5-70fd-4754-a2ed-457d4356558e","resolution":{"observed_at":"2026-08-07T05:19:30.621830Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+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":"2026-08-07T05:19:30.054145Z","title":"Direct preference optimization: Your language model is secretly a reward model.Advances in Neural Information Processing Systems, 36:53728–53741, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.054145Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:fe13bd4278291933bccff85a5d625fb9138722ccf88e4445ecfc01bdd6584699","observation_id":"ed5824e2-6c5d-4a93-87b9-e9df474926b4","resolution":{"observed_at":"2026-08-07T05:19:30.054145Z","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":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-07T05:19:30.057896Z","title":"Locating and editing factual associations in gpt.Advances in neural information processing systems, 35:17359–17372, 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.057896Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:010fd3729506ac7465c8cb0f4d3edde6e80f3ad058e55daab5977695d9c17680","observation_id":"45b52d2d-15c8-471a-b2de-8ce9a1f802ea","resolution":{"observed_at":"2026-08-07T05:19:30.057896Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2409.05907","last_updated":"2025-02-17T20:23:19Z","snapshot_observed_at":"2026-08-04T04:21:28.594163Z","submitted_at":"2024-09-06T15:47:40Z","title":"Programming Refusal with Conditional Activation Steering","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2409.05907","snapshot_observed_at":"2026-08-07T05:19:30.061438Z","title":"Programming refusal with conditional activation steering.arXiv preprint arXiv:2409.05907, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.061438Z"},"links":{"cited_paper":"/paper/2409.05907","citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:0869356bbafd39aa07694b20cc084789c78a5850b306bb27509527d06e8f87a6","observation_id":"7eb8e8d5-2c91-46c9-9fc7-679ea25f9da8","resolution":{"observed_at":"2026-08-07T05:19:30.061438Z","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":"2026-08-07T05:19:30.598442Z","title":"TruthX: Alleviating hallucinations by editing large language models in truthful space","venue":null,"work_id":"bf19c517-1367-4932-a76d-b56c71a7f880","year":2024},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.065680Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:a10f62bff0ad07834f9353d2fb0b7e0ecf391c3036e7174f676bc407e5394414","observation_id":"cba53e92-636d-477b-a5d3-569ca5ec26c3","resolution":{"observed_at":"2026-08-07T05:19:30.601859Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2406.11717","last_updated":"2024-10-30T18:57:07Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2024-06-17T16:36:12Z","title":"Refusal in Language Models Is Mediated by a Single Direction","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2406.11717","snapshot_observed_at":"2026-08-07T05:19:30.068919Z","title":"Refusal in language models is mediated by a single direction.arXiv preprint arXiv:2406.11717, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.068919Z"},"links":{"cited_paper":"/paper/2406.11717","citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:daee1d8735f7853d4574f151e29e716ff651c6bc4a324892af9098a8031e0a68","observation_id":"0526728d-d7ea-4bc2-84b4-d3476aacf0b6","resolution":{"observed_at":"2026-08-07T05:19:30.068919Z","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":"2026-08-07T05:19:30.588534Z","title":"Truthflow: Truthful LLM generation via representation flow correction","venue":null,"work_id":"c599ac06-f849-470d-9b8f-efb519a66460","year":2025},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.072453Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:cf24c579c54387819f4cc5aaa5d7ffbafedc5082b32819cbe02b521527ba5cd8","observation_id":"05f4ba7b-2112-4b02-ac9f-af9092261ee9","resolution":{"observed_at":"2026-08-07T05:19:30.592290Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+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":"2026-08-07T05:19:30.579565Z","title":null,"venue":null,"work_id":"5fef0164-6eec-46f2-a4eb-259442534413","year":2024},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.075628Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:e840bc159527b882de0b0ff87c0489ffac1555247335d344588d9e44d9aef2af","observation_id":"3237afe8-e269-4558-9617-b3417eee2b73","resolution":{"observed_at":"2026-08-07T05:19:30.582773Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+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":"2026-08-07T05:19:30.569644Z","title":"Steering knowledge selection behaviours in LLMs via SAE-based representation engineering","venue":null,"work_id":"0891ebec-ee60-4cce-878d-5861e1d16525","year":2025},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.079688Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:84638f309248f1de5eedd975a17131077b08f96bf8861515f0ecd1d535ccc2cc","observation_id":"6120db2a-46b1-4c39-856c-b53628a09f36","resolution":{"observed_at":"2026-08-07T05:19:30.573045Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+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":"2026-08-07T05:19:30.560757Z","title":"Lofit: Localized fine-tuning on LLM representations","venue":null,"work_id":"cd8cd4a0-6857-4bec-87c8-36fd397c1bcb","year":2024},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.083745Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:a2891e92b7a150fb3992f401762628e537bbe6489bc6318e68606a7da7f56c7f","observation_id":"13d1151f-dd77-47e0-b3f2-ffff6f123dca","resolution":{"observed_at":"2026-08-07T05:19:30.564081Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+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":"2026-08-07T05:19:30.550699Z","title":"Induction heads as an essential mechanism for pattern matching in in-context learning","venue":null,"work_id":"821f2ed2-8906-491a-a653-152f521ceb13","year":2025},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.086902Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:a8a6d122b6067d46557de2f770506a4a86bef08cb6ee63f423eae6347f3535fa","observation_id":"6d01e4b4-72b9-435b-995e-69b72b3b691f","resolution":{"observed_at":"2026-08-07T05:19:30.554196Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2209.11895","last_updated":"2022-09-24T00:43:19Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2022-09-24T00:43:19Z","title":"In-context Learning and Induction Heads","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2209.11895","snapshot_observed_at":"2026-08-07T05:19:30.090177Z","title":"In-context learning and induction heads.arXiv preprint arXiv:2209.11895, 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.090177Z"},"links":{"cited_paper":"/paper/2209.11895","citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:8c7df67b30566d0db4317c45d81b36ac35a2dc57cb5544d26bf7532f15c0d848","observation_id":"ff1fb717-3da4-45ec-9b50-2b6dd27db6f9","resolution":{"observed_at":"2026-08-07T05:19:30.090177Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2404.15574","last_updated":"2024-04-24T00:24:03Z","snapshot_observed_at":"2026-08-06T03:07:12.026697Z","submitted_at":"2024-04-24T00:24:03Z","title":"Retrieval Head Mechanistically Explains Long-Context Factuality","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2404.15574","snapshot_observed_at":"2026-08-07T05:19:30.093604Z","title":"Retrieval head mechanistically explains long-context factuality.arXiv preprint arXiv:2404.15574, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.093604Z"},"links":{"cited_paper":"/paper/2404.15574","citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:9f9ba0ceeee78b554f2a7bde56f7f691202964e6d4658fd3bbd96fc16bceccf3","observation_id":"1928aaf2-3c6e-47ef-b15e-9150072269f2","resolution":{"observed_at":"2026-08-07T05:19:30.093604Z","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":"2026-08-07T05:19:30.539718Z","title":"A glitch in the matrix? locating and detecting language model grounding with fakepedia","venue":null,"work_id":"7e29631e-d56c-4675-8ddf-826a459411fb","year":2024},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.096868Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:a7c6ff26ced3fdbc3b3390336e277dcbd2244fc379d525ebe14c89deb37e396c","observation_id":"084a692f-43d5-4ba7-bf6c-d75f655a7825","resolution":{"observed_at":"2026-08-07T05:19:30.543334Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2405.00208","last_updated":"2024-10-13T21:05:31Z","snapshot_observed_at":"2026-08-07T14:00:54.767580Z","submitted_at":"2024-04-30T21:20:17Z","title":"A Primer on the Inner Workings of Transformer-based Language Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2405.00208","snapshot_observed_at":"2026-08-07T05:19:30.100608Z","title":"A primer on the inner workings of transformer-based language models.arXiv preprint arXiv:2405.00208, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.100608Z"},"links":{"cited_paper":"/paper/2405.00208","citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:b7ac06d45924ee433a9548a3a747433adc28645ecb7e4239d759b2512bb95c29","observation_id":"a7746e88-e267-4a16-ab21-a2c6f35cccb0","resolution":{"observed_at":"2026-08-07T05:19:30.100608Z","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":"2026-08-07T05:19:30.530060Z","title":"Truthfulqa: Measuring how models mimic human falsehoods","venue":null,"work_id":"88956339-0fd3-42af-9b5d-b58facdff469","year":2022},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.107445Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:94841adfe63fd02b7a50dfc42bc1ff36da815ee35f2ac785b31893df1fb090e1","observation_id":"b648afde-bc63-4a18-a04f-c5e3d744c08e","resolution":{"observed_at":"2026-08-07T05:19:30.533886Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+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":"2026-08-07T05:19:30.520080Z","title":"MQuAKE: Assessing knowledge editing in language models via multi-hop questions","venue":null,"work_id":"8ffb67ad-fa52-4b4b-adaf-22595697ef2b","year":2023},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.110855Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:e7b6e4b910fe4775fb18a8fbb39d75689c66eac62a6c378deb00edcca9661206","observation_id":"0abfeed8-dfdd-465f-a77a-1a7a6c0ea2b7","resolution":{"observed_at":"2026-08-07T05:19:30.523663Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+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":"2026-08-07T05:19:30.509649Z","title":"Hamilton","venue":null,"work_id":"0e395fa0-a838-4902-9459-84926a89d185","year":2019},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.114559Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:4db07cea193405346f38c206cc59f785342232aecb230a22a76d630fe2a7bf07","observation_id":"6c06eb5b-596e-400d-a1a8-e4daf14066b7","resolution":{"observed_at":"2026-08-07T05:19:30.513454Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+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":"2026-08-07T05:19:30.500261Z","title":"Entity-based knowledge conflicts in question answering","venue":null,"work_id":"f5ec063c-35e0-4364-ad4d-382fe6bdb79d","year":2021},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.117429Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:3495189ac9fbee61a1410aa7e7937ff5aa5fbb090a4d5472a134dbbe67acaaaa","observation_id":"31f0770d-b65e-4a5d-8f75-1c84b5e6e8ed","resolution":{"observed_at":"2026-08-07T05:19:30.503154Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+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":"2026-08-07T05:19:30.485470Z","title":"Why so gullible? enhancing the robustness of retrieval-augmented models against counterfactual noise","venue":null,"work_id":"b3e0715c-f2d9-4b77-a75a-a2e52852f872","year":2024},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.124755Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:72fb46551602f06775a059978bdffc1045baf413e43ffc431d863babd3c89b59","observation_id":"a92fb971-1b28-41cb-9d9a-f24a910c4efb","resolution":{"observed_at":"2026-08-07T05:19:30.488851Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2212.09251","last_updated":"2022-12-19T05:13:52Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2022-12-19T05:13:52Z","title":"Discovering Language Model Behaviors with Model-Written Evaluations","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2212.09251","snapshot_observed_at":"2026-08-07T05:19:30.128361Z","title":"Discovering language model behaviors with model-written evaluations.arXiv preprint arXiv:2212.09251, 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.128361Z"},"links":{"cited_paper":"/paper/2212.09251","citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:33f8661665a7fa30d1725e0712b7eb24f973c0303ecbf9036e5a829dc39194c0","observation_id":"f1a68ef2-c266-4c78-bef4-a61e40d18845","resolution":{"observed_at":"2026-08-07T05:19:30.128361Z","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":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-07T05:19:30.131991Z","title":"Contrastive decoding: Open-ended text generation as optimization","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.131991Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:8094b90f2c9be726c34f9ce4dd1b1a2b1f028dc89ada6bbe098d7bdfca373f16","observation_id":"1aaa82a5-8eba-468f-a660-25793b5c8297","resolution":{"observed_at":"2026-08-07T05:19:30.131991Z","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":"2026-08-07T05:19:30.468797Z","title":"Glass, and Pengcheng He","venue":null,"work_id":"15a492ae-d865-4e45-9526-26471c911706","year":2024},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.134963Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:9a65eef764183ffeb5764804e0c29ac37ec6635fb8a4b3d4a6a46964c257adca","observation_id":"71c95c59-7702-4b58-b4d8-9628968457c2","resolution":{"observed_at":"2026-08-07T05:19:30.472909Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+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":"2026-08-07T05:19:30.459119Z","title":"In- context sharpness as alerts: An inner representation perspective for hallucination mitigation","venue":null,"work_id":"4de787b5-2df7-4745-9af6-875451b0d4f4","year":2024},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.138389Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:b3d711e1b84a4a93a9a5ca5de79a079f07c1a8dbba91f532f87ac1589676b7b0","observation_id":"4a1de50c-71c6-4266-b82d-a2c93b770081","resolution":{"observed_at":"2026-08-07T05:19:30.462691Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+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":"2026-08-07T05:19:30.448887Z","title":"In-context learning creates task vectors","venue":null,"work_id":"9caeec36-8866-401c-bf20-8c79fdddb7fc","year":2023},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.142058Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:005dcf5cc21999cca36595e1d9482c555ff5435dbf6446a73be5633bc48e03cc","observation_id":"134168d1-0bf8-4bdd-9b67-257b4d762f08","resolution":{"observed_at":"2026-08-07T05:19:30.452959Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2308.10248","last_updated":"2024-10-10T13:20:13Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2023-08-20T12:21:05Z","title":"Steering Language Models With Activation Engineering","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2308.10248","snapshot_observed_at":"2026-08-07T05:19:30.145386Z","title":"Activation addition: Steering language models without optimization.CoRR, abs/2308.10248, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.145386Z"},"links":{"cited_paper":"/paper/2308.10248","citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:e9d1d5aa93b5fe0e58f57aac3eded1568f5ad8febb1cc48148cde873c004c28d","observation_id":"8211caa3-5d71-4ef9-93d1-a9d49dec6726","resolution":{"observed_at":"2026-08-07T05:19:30.145386Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2405.09719","last_updated":"2024-11-03T11:12:14Z","snapshot_observed_at":"2026-08-05T15:48:52.479297Z","submitted_at":"2024-05-15T22:28:23Z","title":"Spectral Editing of Activations for Large Language Model Alignment","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2405.09719","snapshot_observed_at":"2026-08-07T05:19:30.148340Z","title":"Ponti, and Shay B","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.148340Z"},"links":{"cited_paper":"/paper/2405.09719","citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:0a6e39417c4647f4928932d2ba869cf18080bde468754695f73abb53ff7c0181","observation_id":"d72fe0a3-2506-435a-b70c-91bbfadbd3ce","resolution":{"observed_at":"2026-08-07T05:19:30.148340Z","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":"2026-08-07T05:19:30.438708Z","title":"Trusting your evidence: Hallucinate less with context-aware decoding","venue":null,"work_id":"0890670a-d3b3-4f13-bd85-1d06bc35ad44","year":2024},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.151832Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:3f27c92e2fddff217173adf665e41657eaa41c4e5521d3ff209ecfb041e02ce6","observation_id":"2a57ad75-c2ad-4fbc-a97a-d6b6733aff18","resolution":{"observed_at":"2026-08-07T05:19:30.442153Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2005.14165","last_updated":"2020-07-22T19:47:17Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2020-05-28T17:29:03Z","title":"Language Models are Few-Shot Learners","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2005.14165","snapshot_observed_at":"2026-08-07T05:19:30.155432Z","title":"Language models are few-shot learners.arXiv preprint arXiv:2005.14165, 2020","venue":null,"work_id":null,"year":2005},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.155432Z"},"links":{"cited_paper":"/paper/2005.14165","citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:adbe48f6184930118740c2f3270556a53e6c55f872f460f01f2d20aff44fe113","observation_id":"6a2a0585-b7ee-42d5-9638-b634c8a7e74f","resolution":{"observed_at":"2026-08-07T05:19:30.155432Z","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":"2026-08-07T05:19:30.428790Z","title":"GQA: training generalized multi-query transformer models from multi-head checkpoints","venue":null,"work_id":"edab3acb-9f43-46bd-ab5c-76f2504302e8","year":2023},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.158647Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:44f5ff1d108579f654b9aa3099ab87a24c11e36030f51981c226031628fb1d18","observation_id":"127cda6a-d2c4-4cff-86f1-e70a39549cb9","resolution":{"observed_at":"2026-08-07T05:19:30.432771Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2408.12076","last_updated":"2024-08-22T02:33:13Z","snapshot_observed_at":"2026-07-06T19:04:22.495890Z","submitted_at":"2024-08-22T02:33:13Z","title":"ConflictBank: A Benchmark for Evaluating the Influence of Knowledge Conflicts in LLM","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2408.12076","snapshot_observed_at":"2026-08-07T05:19:30.165086Z","title":"Conflictbank: A benchmark for evaluating the influence of knowledge conflicts in llm.arXiv preprint arXiv:2408.12076, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.165086Z"},"links":{"cited_paper":"/paper/2408.12076","citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:fb90e5a1ef81d895e09a895d497737dba87dfc2cb832d4b92a9241928095b179","observation_id":"53720205-313c-49ba-bb67-dfd2fef56f4e","resolution":{"observed_at":"2026-08-07T05:19:30.165086Z","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":"2026-08-07T05:19:30.413829Z","title":"Adaptive chameleon or stubborn sloth: Revealing the behavior of large language models in knowledge conflicts","venue":null,"work_id":"8d314b59-c900-4a5a-901a-13248135d742","year":2024},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.169360Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:6b0d5eefb68a46acc7df9f654a94f36e50ed68f13e5f94e39ab4c6f8b1079bf0","observation_id":"597b7251-8dc8-4e8b-a28f-977889e3539f","resolution":{"observed_at":"2026-08-07T05:19:30.417093Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+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":"2026-08-07T05:19:30.403696Z","title":"Combating misinformation in the age of llms: Opportunities and challenges.AI Magazine, 2023","venue":null,"work_id":"213a7353-bfea-498a-bcf8-5efcdb014dcf","year":2023},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.173290Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:e0445aef36225d7dd2bc61945eb0fd27bdfb9a8a74f530daac822f52db6539ff","observation_id":"1d86a6e5-824b-40d0-bb41-d85bf2ce674e","resolution":{"observed_at":"2026-08-07T05:19:30.407836Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+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":"2026-08-07T05:19:30.393170Z","title":"When not to trust language models: Investigating effectiveness of parametric and non-parametric memories","venue":null,"work_id":"e093426a-4dfc-437a-a2dc-abd965c8a72e","year":2023},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.176615Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:7e96d3b5c0d5f542e8c79df0d577624c7f388465235634f05d6d025fa2ddfe68","observation_id":"229db62a-32f6-4090-99d9-179127be7c88","resolution":{"observed_at":"2026-08-07T05:19:30.396450Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+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":"2026-08-07T05:19:30.180154Z","title":"Pytorch: An imperative style, high-performance deep learning library.Advances in neural information processing systems, 32, 2019","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.180154Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:2b882873fc46136045846c9f0f5fd3a4be830db10b115c19e6930c250a783494","observation_id":"53dc8559-cf7b-4154-9f91-b3e70cfab6d4","resolution":{"observed_at":"2026-08-07T05:19:30.180154Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1910.03771","last_updated":"2020-07-14T03:42:34Z","snapshot_observed_at":"2026-07-06T08:27:58.343233Z","submitted_at":"2019-10-09T03:23:22Z","title":"HuggingFace's Transformers: State-of-the-art Natural Language Processing","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1910.03771","snapshot_observed_at":"2026-08-07T05:19:30.184036Z","title":"Huggingface’s transformers: State-of-the-art natural language processing.arXiv preprint arXiv:1910.03771, 2019","venue":null,"work_id":null,"year":1910},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.184036Z"},"links":{"cited_paper":"/paper/1910.03771","citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:a16ef5e0f8303f8cae516f082b4b81c538b45a98f8e754fea06af3a52716692a","observation_id":"2fda548e-59aa-4c2c-bfda-0f59f7b8db89","resolution":{"observed_at":"2026-08-07T05:19:30.184036Z","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":"2026-08-07T05:19:30.375322Z","title":"Evaluating the ripple effects of knowledge editing in language models.Trans","venue":null,"work_id":"caea10bf-353e-4547-8201-b3d51d34ab8f","year":2024},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.187629Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:19940f5090babe68dea4b1c1a53034b1ad68860d61bfcb082338eac06976a83d","observation_id":"12733733-98e5-4040-b41c-88490cb3007b","resolution":{"observed_at":"2026-08-07T05:19:30.379182Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+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":"2026-08-07T05:19:30.364995Z","title":"Advanced AI Risk","venue":null,"work_id":"bf2d7705-1475-4b10-b466-6ec4d1989186","year":2014},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.190619Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:5ff7a4502deda98139dd6989e5f353f824ad92f8c4530260b2393fb81dc17c11","observation_id":"0799f130-07aa-4ae3-a0b6-9974aeee7b32","resolution":{"observed_at":"2026-08-07T05:19:30.368957Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+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":"2026-08-07T05:19:30.355137Z","title":null,"venue":null,"work_id":"2c2cbc09-ef7e-499a-ac46-1f0e3a1e03d2","year":null},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.194615Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:5567c04062f720216d8051a20207c49938a5a7f0ff191ef58f863b35a223e98f","observation_id":"bc330a0d-1e46-4e88-9db5-514e971cad13","resolution":{"observed_at":"2026-08-07T05:19:30.358220Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+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":"2026-08-07T05:19:30.344030Z","title":"Typically, τ is computed as the (100−p) -th percentile over the collection of all head scores, we useτ= 5for all experiments: τ= Percentile {sl,h}l,h,100−p","venue":null,"work_id":"77790b36-89bd-4816-a538-e5d7958fc207","year":null},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.198500Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:a376cb6b064278f7a2614d87c67992bf0e06e1e14962d0e4e9e8dd54a71bdf8c","observation_id":"41100c45-e8f1-4482-aed8-33f29049a1ba","resolution":{"observed_at":"2026-08-07T05:19:30.348885Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+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":"2026-08-07T05:19:30.332621Z","title":"linear” and “SqExp","venue":null,"work_id":"b67701c3-940d-4838-81c8-7abe692461c0","year":2003},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.202178Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:50025938eb2e744f54e4160aef98f4e9151048c3c90d75796fd873fba1b4fda2","observation_id":"9f8a16d0-61a2-461c-8a4b-e34e0c2ddf9d","resolution":{"observed_at":"2026-08-07T05:19:30.337176Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+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":"2026-08-07T05:19:30.120899Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":2021,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.120899Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:6726cbedeb69149fb60339747e4be6799037691fcff8d125346ddfed42dfa02f","observation_id":"2b66d5ea-41a7-46b7-9b4b-64f46521bc7c","resolution":{"observed_at":"2026-08-07T05:19:30.120899Z","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":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-07T05:19:30.161792Z","title":null,"venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering","version":3},"reference_index":4901,"source":"pdf_text","source_observed_at":"2026-08-07T05:19:30.161792Z"},"links":{"citing_paper":"/paper/2506.08359"},"observation_digest":"sha256:c01592ce8cca0836e9e7d05a364d869f72bc7ae341edddd83b0ecb49ea1a1e29","observation_id":"30b7d024-9f38-4efa-a940-c7279563ef91","resolution":{"observed_at":"2026-08-07T05:19:30.161792Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2506.08359","last_updated":"2026-07-13T06:33:40Z","latest_version":3,"primary_category":"cs.CL","snapshot_observed_at":"2026-08-08T00:02:40.624808Z","submitted_at":"2025-06-10T02:16:50Z","title":"REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering"},"reference_resolution":{"displayed":45,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":19,"verified_exact":0,"verified_fuzzy":26},"total_outbound_references":45},"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":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+00:00","source":"retraction_watch"}],"thesis":"As of 10 August 2026, this Paper Citation Record lists 45 of 45 outbound references and 0 inbound Pith citation observations for arXiv:2506.08359."}