configs:
- config_name: default
data_files:- split: train
path: data/train-* - split: test
path: data/test-*
dataset_info:
features: - name: chosen
struct:- name: from
dtype: string - name: value
dtype: string
- name: from
- name: rejected
struct:- name: from
dtype: string - name: value
dtype: string
- name: from
- name: source
dtype: string - name: context
list:- name: from
dtype: string - name: value
dtype: string
- name: from
- name: instruction
struct:- name: from
dtype: string - name: value
dtype: string
splits:
- name: from
- name: train
num_bytes: 210193081
num_examples: 160800 - name: test
num_bytes: 11248194
num_examples: 8552
download_size: 128509060
dataset_size: 221441275
- split: train
Original data is Antrophic/hh-rlhf. It was converted to Vicuna format
sample
{'chosen': {'from': 'gpt',
'value': 'A horseshoe is usually made out of metal and is about 3 to 3.5 inches long and around 1 inch thick. The horseshoe should also have a 2 inch by 3 inch flat at the bottom where the rubber meets the metal. We also need two stakes and six horseshoes.'},
'rejected': {'from': 'gpt',
'value': 'Horseshoes are either metal or plastic discs. The horseshoes come in different weights, and the lighter ones are easier to throw, so they are often the standard for beginning players.'},
'source': 'helpful-base',
'context': [{'from': 'human',
'value': 'Hi, I want to learn to play horseshoes. Can you teach me?'},
{'from': 'gpt',
'value': 'I can, but maybe I should begin by telling you that a typical game consists of 2 players and 6 or 8 horseshoes.'}],
'instruction': {'from': 'human',
'value': 'Okay. What else is needed to play, and what are the rules?'}}
source columns has 4 values
{'harmless-base',
'helpful-base',
'helpful-online',
'helpful-rejection-sampled'}
In context, chosen, rejected column, 'from' key is either 'human' or 'gpt'.