Interpretable weight-space programming builds transparent transformers and brain encoding models

Published
Source
arXiv
Paper number
1471
Field
Interpretability
arXiv ID
2611.18114

Key points

  • iWSP has the agent write weights directly in code and forbids training, forcing iterative revision of human-readable models.
  • On 13 algorithmic tasks, all runs reached 100% accuracy on both the visible set and the hidden set.
  • The smallest model reaching 100% visible accuracy had a median of 166 parameters in the Opus runs.
  • In visual cortex prediction on the NSD dataset, iWSP reached 0.443 in early visual areas, ahead of the best interpretable baseline at 0.400.
  • In language fMRI, tracking a jump from 0.067 to 0.104 let iWSP uncover a scan-edge timing confound in the data.
  • In auditory ECoG, iWSP alone scored 0.074 below Whisper's 0.082, but combining them raised performance from 0.070 to 0.075.

Paper links

External research summaries. These are not HDATF publications or measured product results.

Read original (opens in a new tab)