metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: completion
dtype: string
splits:
- name: train
num_bytes: 1197643690
num_examples: 277573
- name: validation
num_bytes: 63630789
num_examples: 14610
- name: test
num_bytes: 65536728
num_examples: 15379
download_size: 820321648
dataset_size: 1326811207
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*