metadata
dataset_info:
features:
- name: file_name
dtype: string
- name: comments
dtype: string
- name: code_string
dtype: string
- name: __index_level_0__
dtype: int64
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
- name: labels
sequence: int64
splits:
- name: train
num_bytes: 46347008721
num_examples: 1359405
download_size: 1838655465
dataset_size: 46347008721
configs:
- config_name: default
data_files:
- split: train
path: data/train-*