metadata
dataset_info:
features:
- name: context
dtype: string
- name: target
dtype: string
splits:
- name: train_fewshots
num_bytes: 45045
num_examples: 100
- name: train_ICL
num_bytes: 3023334
num_examples: 7000
- name: train_large
num_bytes: 95075188
num_examples: 219303
- name: val
num_bytes: 425769
num_examples: 1000
- name: test_large
num_bytes: 11299052
num_examples: 26556
download_size: 39589753
dataset_size: 109868388
configs:
- config_name: default
data_files:
- split: train_fewshots
path: data/train_fewshots-*
- split: train_ICL
path: data/train_ICL-*
- split: train_large
path: data/train_large-*
- split: val
path: data/val-*
- split: test_large
path: data/test_large-*