metadata
dataset_info:
features:
- name: cumulative_reward
sequence: float64
- name: input_ids
sequence: int64
- name: attention_mask
sequence: int64
- name: action_ids_mask
sequence: bool
- name: end_ids_mask
sequence: bool
- name: env_id
sequence: int64
- name: labels
sequence: int64
splits:
- name: test
num_bytes: 7420140
num_examples: 45
- name: train
num_bytes: 1590878016
num_examples: 9648
download_size: 16298787
dataset_size: 1598298156
configs:
- config_name: default
data_files:
- split: test
path: data/test-*
- split: train
path: data/train-*