metadata
dataset_info:
features:
- name: blob_id
dtype: string
- name: formatted
dtype: string
- name: input_ids
list: int32
- name: tokens
list: string
- name: rows
list: int64
- name: cols
list: int64
splits:
- name: train
num_bytes: 59164561
num_examples: 900
download_size: 9472840
dataset_size: 59164561
configs:
- config_name: default
data_files:
- split: train
path: data/train-*