pith. machine review for the scientific record.
sign in

Towards automated circuit discovery for mechanistic interpretability

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

fields

cs.AI 1

years

2025 1

verdicts

UNVERDICTED 1

representative citing papers

Thinking Sparks!: Emergent Attention Heads in Reasoning Models During Post Training

cs.AI · 2025-09-30 · unverdicted · novelty 6.0

Post-training on reasoning tasks sparks the emergence of specialized attention heads that enable structured computation, with SFT adding stable heads while GRPO uses dynamic activation and pruning tied to reward signals, and controllable think models relying on compensatory heads instead of specific

citing papers explorer

Showing 1 of 1 citing paper.

  • Thinking Sparks!: Emergent Attention Heads in Reasoning Models During Post Training cs.AI · 2025-09-30 · unverdicted · none · ref 11

    Post-training on reasoning tasks sparks the emergence of specialized attention heads that enable structured computation, with SFT adding stable heads while GRPO uses dynamic activation and pruning tied to reward signals, and controllable think models relying on compensatory heads instead of specific