{"type":"post","id":"goodfire-introduces-mechanistic-shaping-adding-interpretability-based-penalties-09f568","slug":"goodfire-introduces-mechanistic-shaping-adding-interpretability-based-penalties-09f568","title":"Goodfire introduces mechanistic shaping, adding interpretability-based penalties to the training loss.","description":"Goodfire’s November 5, 2026 research post describes mechanistic shaping: adding penalties derived from a model’s internal representations to steer how it solves a task, not merely whether it returns the desired answer.","retrieval_nugget":"Goodfire’s November 5, 2026 research post describes mechanistic shaping: adding penalties derived from a model’s internal representations to steer how it solves a task, not merely whether it returns the desired answer. The reported experiments target misleading shortcuts, sycophancy, and reward hacking while testing whether the targeted concepts remain monitorable after training.","published_at":"2026-11-06","updated_at":"2026-11-07","record_date":"2026-11-06","date_kind":"published_at","topics":["models"],"entities":["Goodfire","Mechanistic shaping"],"related_patterns":[],"novelty_label":"new_fact","what_changed":"Goodfire’s new research introduces mechanistic shaping, adding internal-representation penalties to training so models are rewarded for reaching answers for the intended reasons.","prior_refs":[],"source_url":"https://www.goodfire.com/research/mechanistic-shaping","source_urls":["https://www.goodfire.com/research/mechanistic-shaping"],"schema_version":"newruntime-agent-readable-v0.2","stable_id":"post:goodfire-introduces-mechanistic-shaping-adding-interpretability-based-penalties-09f568","status":"published","visuals":[],"editor":{"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]},"routes":{"html":"https://newruntime.com/posts/goodfire-introduces-mechanistic-shaping-adding-interpretability-based-penalties-09f568/","markdown":"https://newruntime.com/posts/goodfire-introduces-mechanistic-shaping-adding-interpretability-based-penalties-09f568.md","json":"https://newruntime.com/posts/goodfire-introduces-mechanistic-shaping-adding-interpretability-based-penalties-09f568.json"},"markdown_url":"https://newruntime.com/posts/goodfire-introduces-mechanistic-shaping-adding-interpretability-based-penalties-09f568.md","next_reads":[{"type":"topic","path":"/topics/models/","url":"https://newruntime.com/topics/models/","title":"Models","reason":"Explore the models topic hub.","media_type":"text/html","markdown_url":"https://newruntime.com/topics/models.md","json_url":"https://newruntime.com/topics/models.json"},{"type":"collection","path":"/posts/","url":"https://newruntime.com/posts/","title":"Field Notes","reason":"Open the latest editorial analysis.","media_type":"text/html","json_url":"https://newruntime.com/posts.json"},{"type":"collection","path":"/sources/","url":"https://newruntime.com/sources/","title":"Sources","reason":"See the original sources this archive cites.","media_type":"text/html"}]}