finding
active
finding:more-than-95-of-r1-s-english-reasoning-traces-begin-with-okayMore than 95% of R1's English reasoning traces begin with 'Okay,'
Evidence for the training-prefix hypothesis.
Neighborhood — ranked by edge-count
Papers (1)
paper
Hypotheses (1)
hypothesis
- Explanation for late steering and attention sinks.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Central interpretive assertion of the post.
- Attention sinks in R1 appear toward the end of the response prefix rather than at the start of responsesfinding0.778Suggests R1 does not model its true response as begun until after the prefix.
- First steering insight specific to R1.
- Suggestive evidence for language-independent truth representation in LLMs
- R1 implicitly recognizes confusion or incoherence when activations are overly perturbed, causing it to stop and course-correcthypothesis0.758Working hypothesis for the oversteering reversion.
- R1 general reasoning SAE learned features reconstructing reasoning behaviors such as backtrackingfinding0.754Curated features show core reasoning behaviors are captured.
- Demonstrates NLAs' ability to surface hypotheses that lead to discovery of root cause (malformed training data).
- Evidence that reasoning length does not track safety performance