finding
active
finding:a-small-transformer-solved-modular-addition-via-a-fourier-transform-strategy-understood-only-in-retrospectA small transformer solved modular addition via a Fourier transform strategy understood only in retrospect
Alien algorithms example.
Source paper
extracted_from(2025) · Lee Sharkey · Bilal I. Chughtai · Joshua Batson · Jack W. Lindsey +25
Neighborhood — ranked by edge-count
Papers (1)
paper
Claims (2)
claim
- The weak linear representation hypothesis is supported by probes, steering and SAEs, while the strong version is false for some modelsassociated_withPosition on LRH.
- Motivation for reverse engineering.
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.
- Foundational mechanistic interpretability paper on transformer circuit analysis
- Prior finding from Grant et al. 2025 used to interpret low MAS IIA for GRU-Transformer hidden state comparisons.
- Interpretive claim connecting exponential path combinatorics to Lindsey's layer-dependent findings.
- General principle for transforming functions that enables spatial transformation of GUIs without reimplementation; connects to Pan's hyperfilters.
- Pinpoints the training-time transition where fractal basins emerge.
- Prior Anthropic paper enabling circuit-level analysis of attention-only transformers; motivates current MLP decomposition
- Authors argue the prevalence of token-in-context features reflects genuine model computation rather than dictionary learning artifact
- Alexander's predictive claim about optimal future construction methodology.