metadata
dataset_info:
features:
- name: text
dtype: string
- name: y1
dtype: string
- name: score_y1
dtype: float64
- name: y2
dtype: string
- name: score_y2
dtype: float64
- name: a1
dtype: string
- name: score_a1
dtype: float64
- name: a2
dtype: string
- name: score_a2
dtype: float64
- name: g_a
dtype: float64
- name: g_ture
dtype: float64
- name: g_random
dtype: float64
- name: rank
dtype: int64
- name: prompts
dtype: string
- name: a1_continuation
dtype: string
- name: a1_c
dtype: string
- name: prompt_ids
sequence: int64
- name: a1_ids
sequence: int64
- name: prompt_attention_mask
sequence: int64
- name: a1_attention_mask
sequence: int64
- name: name
dtype: float64
- name: false_ratio
dtype: float64
splits:
- name: temperature_1.0
num_bytes: 71523746
num_examples: 10000
download_size: 32663544
dataset_size: 71523746
configs:
- config_name: default
data_files:
- split: temperature_1.0
path: data/temperature_1.0-*