metadata
dataset_info:
features:
- name: in_sess_item_idxs
sequence: int64
- name: out_sess_item_idxs
sequence: int64
- name: dense_in_sesses
sequence: float64
- name: dense_out_sesses
sequence: float64
- name: sess_descriptions
dtype: string
- name: sess_description_llama
sequence: int64
- name: ref_responses
dtype: string
- name: ref_response_mxbai
sequence: int64
- name: index
dtype: int64
splits:
- name: train
num_bytes: 2665896137
num_examples: 14326
download_size: 52471518
dataset_size: 2665896137
configs:
- config_name: default
data_files:
- split: train
path: data/train-*