CausalGym covers only linguistic tasks; benchmarking interpretability methods on non-linguistic behaviours remains open

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

• Multi-dimensional linear and non-linear interpretability methods have not been benchmarked on CausalGymquestion0.841
  Identified gap in benchmark coverage; only 1D linear methods are benchmarked
• CausalGym only includes English data; comparable experiments with other languages might yield substantially different resultsquestion0.841
  Identified limitation/gap calling for cross-lingual extension of CausalGym
• CausalGymframework0.790
  Multi-task benchmark of linguistic behaviours for measuring causal efficacy of interpretability methods, adapted from SyntaxGym
• Task accuracy on CausalGym increases consistently with model scale from 0.62 (14M) to 0.89 (6.9B)finding0.760
  Scaling result showing larger pythia models perform better on CausalGym linguistic tasks
• Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models (Marks et al., 2025)concept0.758
  Cited as enabling precise behavioral control through SAE features, extending the same methodological line
• Causal abstraction is not enough for mechanistic interpretability because it becomes vacuous without assumptions about how models encode informationclaim0.751
  Central thesis of the paper
• The same charitable interpretation must be extended to all systems that display observable response patterns that are consistent with animal cognition, including artificial intelligences, metaplastic materials, and robotic systems.claim0.749
  Call to extend the inference of sentience to non-biological systems as well.
• Behavioural patterns associated with subjective experiences in humans are considered valid for inferring cognition in non-human animals but not in diverse other systems including plants.claim0.748
  The double standard pointed out by S&C and endorsed by the authors.