---
dataset_info:
  features:
  - name: text
    dtype: string
  - name: activations
    sequence: float64
  splits:
  - name: train
    num_bytes: 41656850477.0
    num_examples: 1141728
  download_size: 2924704503
  dataset_size: 41656850477.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---