metadata
dataset_info:
features:
- name: text
dtype: string
- name: hf_score
dtype: int64
- name: fasttext_score_v2
dtype: float64
- name: dolma_score
dtype: float64
splits:
- name: train
num_bytes: 5930108510
num_examples: 1000000
- name: validation
num_bytes: 2791386
num_examples: 500
- name: test
num_bytes: 58798191
num_examples: 10000
download_size: 3193890606
dataset_size: 5991698087
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*