Cites: Grokked Transformers are Implicit Reasoners: A Mechanistic Journey to the Edge of Generalization

4 citation edges inside this catalog. A title match is weaker evidence than an identifier match. Zero entries does not mean no citations on the web.

• Universal Transformers
  arxiv:1807.03819 · evidence: title
• Progress measures for grokking via mechanistic interpretability
  arxiv:2301.05217 · evidence: title
• Physics of Language Models: Part 3.2, Knowledge Manipulation
  arxiv:2309.14402 · evidence: title
• From Explicit CoT to Implicit CoT: Learning to Internalize CoT Step by Step
  arxiv:2405.14838 · evidence: title