metadata
dataset_info:
features:
- name: original_prompt
dtype: string
- name: prompt
dtype: string
- name: question
dtype: string
- name: tool_description
dtype: string
- name: tool_call
dtype: string
- name: original_tool_description
dtype: string
- name: is_correct
dtype: int64
- name: answer
dtype: string
- name: golden_answer
dtype: string
splits:
- name: train
num_bytes: 325440730
num_examples: 13391
download_size: 78804636
dataset_size: 325440730
configs:
- config_name: default
data_files:
- split: train
path: data/train-*