← back to paper

Review history

arxiv: 2605.08504 · 2 revisions

A Single Layer to Explain Them All:Understanding Massive Activations in Large Language Models

1. 2026-05-14 CONDITIONAL MODERATE v0.9.0 novelty 7.0
   54436 ms 5474 in 1604 out 2026-05-14T20:58:37.603735+00:00
2. 2026-05-12 UNVERDICTED LOW v0.9.0 novelty 7.0
   48973 ms 5474 in 1257 out 2026-05-12T02:25:23.836564+00:00