metadata
dataset_info:
- config_name: default
features:
- name: uuid
dtype: string
- name: license
dtype: string
- name: used_in
list: string
- name: reasoning
dtype: string
- name: system
dtype: string
- name: interactions
list:
- name: query
dtype: string
- name: answer
dtype: string
- name: think
dtype: string
splits:
- name: train
num_bytes: 9696441122
num_examples: 929237
download_size: 5468208623
dataset_size: 9696441122
- config_name: scored
features:
- name: uuid
dtype: string
- name: license
dtype: string
- name: used_in
list: string
- name: reasoning
dtype: string
- name: system
dtype: string
- name: interactions
list:
- name: query
dtype: string
- name: answer
dtype: string
- name: think
dtype: string
- name: scores
struct:
- name: memory_score
dtype: int64
- name: instruction_score
dtype: int64
- name: freshness_score
dtype: int64
- name: overall_score
dtype: int64
- name: confidence
dtype: string
- name: issues
list: string
- name: rationale
struct:
- name: memory
dtype: string
- name: instruction
dtype: string
- name: freshness
dtype: string
splits:
- name: train
num_bytes: 4381162508
num_examples: 247968
download_size: 2423667084
dataset_size: 4381162508
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- config_name: scored
data_files:
- split: train
path: scored/train-*