metadata
dataset_info:
features:
- name: density
dtype: string
- name: vocab_diversity
dtype: string
- name: local_vs_global
dtype: string
- name: tags
dtype: string
- name: necessary_context
dtype: string
- name: necessary_condition
dtype: string
- name: success_criterion
dtype: string
- name: human_explanation
dtype: string
- name: feature_grade
dtype: string
- name: neuronpedia_id
dtype: string
- name: index
dtype: int64
splits:
- name: test
num_bytes: 29064
num_examples: 205
download_size: 13293
dataset_size: 29064
configs:
- config_name: default
data_files:
- split: test
path: data/test-*
language:
- en