metadata
dataset_info:
features:
- name: Source_Obj
dtype: string
- name: Target_Obj
dtype: string
- name: Thinking
dtype: string
- name: Object
dtype: string
- name: instruction
dtype: string
- name: solution
sequence:
sequence: int64
- name: Conversation
list:
- name: content
dtype: string
- name: role
dtype: string
splits:
- name: train
num_bytes: 4425895669
num_examples: 300000
download_size: 588119217
dataset_size: 4425895669
configs:
- config_name: default
data_files:
- split: train
path: data/train-*