The Fourier feature periods (2, 5, 10) respect standard base-10 addition structure rather than cyclic concept periodicity

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• Fourier features with period 10 contribute to base-10 sum computation in the 28-neuron clusterfinding0.846
  One of the three base-10 Fourier periods identified in the sparse neuron set
• Fourier featuresconcept0.784
  Features identified in Llama-3.1-8B that compute sums using periods respecting base-10 addition (2, 5, 10) rather than concept-specific periods
• Fourier Features for Numerical Computationconcept0.767
• There appears to be a systematic relationship between the frequency of concepts and the dictionary size needed to resolve features for them.claim0.758
  Feature presence depends on concept frequency in training data, with a threshold scaling inversely with alive features.
• Higher-activating feature intervals are systematically more interpretable than lower-activating intervals in human analysisfinding0.751
  Shows interpretability correlates with activation strength, most model effect comes from high activations
• Llama-3.1-8B uses base-10 addition rather than modular addition to compute cyclic concept sumsfinding0.743
  The central empirical finding that computation does not mirror the circular representational structure
• Disjoint neuron clusters for Fourier periodsconcept0.738
  The 28 identified neurons can be partitioned into disjoint clusters each computing a different Fourier period sum
• Does Llama compute modular addition or base-10 addition for cyclic tasks?question0.734
  The specific computational question the paper resolves empirically