---
dataset_info:
  features:
  - name: text
    dtype: string
  - name: label
    dtype: int64
  - name: probs
    dtype: float64
  - name: metadata
    dtype: string
  - name: input_ids
    sequence: int32
  - name: token_type_ids
    sequence: int8
  - name: attention_mask
    sequence: int8
  - name: length
    dtype: int64
  - name: is_edu
    dtype: int64
  splits:
  - name: train
    num_bytes: 1477545184.695477
    num_examples: 123451
  download_size: 933371976
  dataset_size: 1477545184.695477
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---