---
dataset_info:
  features:
  - name: sequence_id
    dtype: int64
  - name: segment_start
    dtype: int64
  - name: input_ids
    list: int32
  - name: attention_mask
    list: int8
  - name: labels
    dtype: int64
  - name: token_length
    dtype: int64
  - name: unknown_token_count
    dtype: int64
  - name: full_labels
    dtype: int64
  splits:
  - name: train
    num_bytes: 31600803
    num_examples: 6123
  - name: validation
    num_bytes: 3710759
    num_examples: 719
  - name: test
    num_bytes: 33618754
    num_examples: 6514
  download_size: 21545670
  dataset_size: 68930316
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
---