metadata
dataset_info:
features:
- name: callID
dtype: string
- name: recordID
dtype: string
- name: branch
dtype: string
- name: source
dtype: string
- name: bucket
dtype: string
- name: CallOutcome
dtype: string
- name: process
dtype: string
- name: input
dtype: string
- name: thinking
dtype: string
- name: output
dtype: string
- name: system_prompt
dtype: string
- name: generated_instruction
dtype: string
splits:
- name: train
num_bytes: 478242
num_examples: 50
download_size: 109190
dataset_size: 478242
configs:
- config_name: default
data_files:
- split: train
path: data/train-*