Cited by: Progress measures for grokking via mechanistic interpretability

17 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.

• Physics of Language Models: Part 1, Learning Hierarchical Language Structures
  arxiv:2305.13673 · evidence: id
• Grokked Transformers are Implicit Reasoners: A Mechanistic Journey to the Edge of Generalization
  arxiv:2405.15071 · evidence: title
• Grokfast: Accelerated Grokking by Amplifying Slow Gradients
  arxiv:2405.20233 · evidence: id
• Pre-trained Large Language Models Use Fourier Features to Compute Addition
  arxiv:2406.03445 · evidence: id
• Arithmetic Without Algorithms: Language Models Solve Math With a Bag of Heuristics
  arxiv:2410.21272 · evidence: title
• Language Models Use Trigonometry to Do Addition
  arxiv:2502.00873 · evidence: title
• Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach
  arxiv:2502.05171 · evidence: title
• The Lookahead Limitation: Why Multi-Operand Addition is Hard for LLMs
  arxiv:2502.19981 · evidence: id
• How do language models learn facts? Dynamics, curricula and hallucinations
  arxiv:2503.21676 · evidence: title
• Reinforcement Learning for Reasoning in Large Language Models with One Training Example
  arxiv:2504.20571 · evidence: id
• Modular Arithmetic: Language Models Solve Math Digit by Digit
  arxiv:2508.02513 · evidence: id
• Why Can't Transformers Learn Multiplication? Reverse-Engineering Reveals Long-Range Dependency Pitfalls
  arxiv:2510.00184 · evidence: id
• In-Context Algebra
  arxiv:2512.16902 · evidence: title
• Language Models Compare Quantities Using Number-specific and Unit-specific Heuristics
  arxiv:2606.03982 · evidence: title
• Emergent Capabilities Arise Randomly from Learning Sparse Attention Patterns
  arxiv:2606.25010 · evidence: id
• AI Finds A Way
  arxiv:2608.23875 · evidence: title
• An Extremely Opinionated Annotated List of My Favourite Mechanistic Interpretability Papers v2 (Alignment Forum)
  url:alignmentforum.org/posts/nffst5mio7bcaqhpa/an-extremely-opinionated-annotated-list-of-my-favourite-1 · evidence: id