Step 2: tool results

get_claim

{
  "include": [
    "arguments",
    "subclaims",
    "related"
  ],
  "claim_id": "b9d866f3-2bda-456c-a10d-bc16f0450577"
}

{"claim":{"id":"b9d866f3-2bda-456c-a10d-bc16f0450577","kind":"claim","canonical_form":"Delta-Crosscoder identifies and mitigates narrow fine-tuning behaviors, outperforming SAE baselines and matching non-SAE methods.","claim_type":"empirical_derived","state":"active","decomposition_status":"complete","importance":0.45,"contestation":0.35,"domains":[],"tags":[{"slug":"mechanistic-interpretability","name":"Mechanistic Interpretability"},{"slug":"large-language-models","name":"Large language models"}],"steward_state":"running","stewarded_at":"2026-11-15T07:19:01.678Z","canonical_direction_note":"The source advances Delta-Crosscoder's claimed ability to identify and mitigate fine-tuning behaviors and its comparative performance over SAE baselines, so the affirmative method-performance thesis is the natural direction.","created_at":"2026-11-13T00:28:42.208Z","updated_at":"2026-11-15T07:15:45.715Z","page_url":"https://minerval.ai/claims/b9d866f3-2bda-456c-a10d-bc16f0450577"},"assessment":null,"coverage":{"examined":false,"on":[],"note":"No judgment of this node is in force. That is not a finding either way."},"quantity":null,"threshold_of":null,"question":null,"formalization":null,"formalization_pending":null,"formalization_history":[],"verification":null,"lean_checks":[],"lean_checks_total":0,"waiting_messages":[{"trigger":"structure_and_assess","from":"the system","sender_node_id":null,"sent_at":"2026-11-13T00:28:42.799Z","paid":false,"summary":"New claim onboarded. Give it the structure and assessment its importance warrants; a minor or settled claim may need no decomposition."},{"trigger":"mandate_plan","from":"a mandate's plan","sender_node_id":null,"sent_at":"2026-11-15T07:15:45.716Z","paid":true,"summary":"First assessment. Read the source reporting this method. Establish: which fine-tuned models and behaviours, what the SAE baselines were and how they were trained, what non-SAE baseline it is said to m…"}],"requests":[],"requests_total":0,"subclaim_count":0,"instance_count":1,"source_map":null,"related":[],"subclaims":[],"subclaims_assessed":0,"arguments":[],"unplaced_argument_passages":[{"instance_id":"cae79856-e5d5-4e85-9253-2d4156b37fdc","source":{"id":"0cd94f15-9fe3-42e7-9c68-8d2b51cb8624","url":"https://arxiv.org/pdf/2603.04426","title":"Delta-Crosscoder: Robust Crosscoder Model Diffing in Narrow Fine-Tuning Regimes"},"verbatim_text":"Delta-Crosscoder matches or exceeds the ability of prior\nSAE-based and non-SAE model diffing methods to identify\nfine-tuning objectives, while substantially reducing analysis\ncomplexity and runtime overhead.","context":"Discussion section 7.2 summarizing comparisons.","stance":"advances","direction":"for"}],"unplaced_argument_passages_total":1}