metadata
dataset_info:
- config_name: drop
features:
- name: messages
list:
- name: content
dtype: string
- name: role
dtype: string
- name: document
dtype: string
- name: answers
sequence: string
- name: shargpt_formatted
list:
- name: from
dtype: string
- name: value
dtype: string
splits:
- name: train
num_bytes: 78867689
num_examples: 29195
download_size: 9598684
dataset_size: 78867689
- config_name: narrativeqa
features:
- name: messages
list:
- name: content
dtype: string
- name: role
dtype: string
- name: document
dtype: string
- name: answers
sequence:
sequence: string
- name: shargpt_formatted
list:
- name: from
dtype: string
- name: value
dtype: string
splits:
- name: train
num_bytes: 284098258
num_examples: 40000
download_size: 10699133
dataset_size: 284098258
- config_name: newsqa
features:
- name: messages
list:
- name: content
dtype: string
- name: role
dtype: string
- name: document
dtype: string
- name: answers
sequence: string
- name: shargpt_formatted
list:
- name: from
dtype: string
- name: value
dtype: string
splits:
- name: train
num_bytes: 573133568
num_examples: 76560
download_size: 71189729
dataset_size: 573133568
- config_name: quoref
features:
- name: answers
sequence: string
- name: document
dtype: string
- name: messages
list:
- name: content
dtype: string
- name: role
dtype: string
- name: shargpt_formatted
list:
- name: from
dtype: string
- name: value
dtype: string
splits:
- name: train
num_bytes: 43068666
num_examples: 10996
download_size: 5976692
dataset_size: 43068666
- config_name: ropes
features:
- name: messages
list:
- name: content
dtype: string
- name: role
dtype: string
- name: document
dtype: string
- name: answers
sequence: string
- name: shargpt_formatted
list:
- name: from
dtype: string
- name: value
dtype: string
splits:
- name: train
num_bytes: 19415418
num_examples: 10924
download_size: 1350788
dataset_size: 19415418
- config_name: sft
features:
- name: messages
list:
- name: content
dtype: string
- name: role
dtype: string
- name: answers
sequence: string
- name: shargpt_formatted
list:
- name: from
dtype: string
- name: value
dtype: string
splits:
- name: train
num_bytes: 666290328
num_examples: 128001
download_size: 385193089
dataset_size: 666290328
- config_name: squad1.1
features:
- name: messages
list:
- name: content
dtype: string
- name: role
dtype: string
- name: document
dtype: string
- name: answers
list:
- name: answer_start
dtype: int64
- name: text
dtype: string
- name: shargpt_formatted
list:
- name: from
dtype: string
- name: value
dtype: string
splits:
- name: train
num_bytes: 231108780
num_examples: 86863
download_size: 25656737
dataset_size: 231108780
- config_name: squad2.0
features:
- name: messages
list:
- name: content
dtype: string
- name: role
dtype: string
- name: document
dtype: string
- name: answers
list:
- name: answer_start
dtype: int64
- name: text
dtype: string
- name: shargpt_formatted
list:
- name: from
dtype: string
- name: value
dtype: string
splits:
- name: train
num_bytes: 232682362
num_examples: 129486
download_size: 28237410
dataset_size: 232682362
- config_name: synthetic_convqa
features:
- name: messages
list:
- name: content
dtype: string
- name: role
dtype: string
- name: document
dtype: string
- name: answers
sequence: string
- name: shargpt_formatted
list:
- name: from
dtype: string
- name: value
dtype: string
splits:
- name: train
num_bytes: 523263735
num_examples: 38689
download_size: 285865309
dataset_size: 523263735
- config_name: tatqa
features:
- name: messages
list:
- name: content
dtype: string
- name: role
dtype: string
- name: document
dtype: string
- name: answers
sequence: string
- name: shargpt_formatted
list:
- name: from
dtype: string
- name: value
dtype: string
splits:
- name: train
num_bytes: 52679111
num_examples: 11501
download_size: 8550834
dataset_size: 52679111
configs:
- config_name: drop
data_files:
- split: train
path: drop/train-*
- config_name: narrativeqa
data_files:
- split: train
path: narrativeqa/train-*
- config_name: newsqa
data_files:
- split: train
path: newsqa/train-*
- config_name: quoref
data_files:
- split: train
path: quoref/train-*
- config_name: ropes
data_files:
- split: train
path: ropes/train-*
- config_name: sft
data_files:
- split: train
path: sft/train-*
- config_name: squad1.1
data_files:
- split: train
path: squad1.1/train-*
- config_name: squad2.0
data_files:
- split: train
path: squad2.0/train-*
- config_name: synthetic_convqa
data_files:
- split: train
path: synthetic_convqa/train-*
- config_name: tatqa
data_files:
- split: train
path: tatqa/train-*