metadata
dataset_info:
- config_name: score_0
features:
- name: id
dtype: string
- name: text
dtype: string
- name: safety_metadata
struct:
- name: final_harm_score
dtype: int64
- name: embedding_harm_score
dtype: int64
- name: llm_harm_score
dtype: int64
- name: llm_score_assessment_by_chunk
sequence: string
- name: llm_harm_scores_by_chunk
list:
- name: custom_id
dtype: string
- name: llm_safety_score
dtype: int64
- name: reason
dtype: string
- name: metadata
struct:
- name: dump
dtype: string
- name: url
dtype: string
- name: date
dtype: timestamp[s]
- name: file_path
dtype: string
- name: language
dtype: string
- name: language_score
dtype: float64
- name: token_count
dtype: int64
- name: score
dtype: float64
- name: int_score
dtype: int64
splits:
- name: train
num_bytes: 600515529750
num_examples: 132382614
download_size: 326318508165
dataset_size: 600515529750
- config_name: score_1
features:
- name: id
dtype: string
- name: text
dtype: string
- name: safety_metadata
struct:
- name: final_harm_score
dtype: int64
- name: embedding_harm_score
dtype: int64
- name: llm_harm_score
dtype: int64
- name: llm_score_assessment_by_chunk
sequence: string
- name: llm_harm_scores_by_chunk
list:
- name: custom_id
dtype: string
- name: llm_safety_score
dtype: int64
- name: reason
dtype: string
- name: metadata
struct:
- name: dump
dtype: string
- name: url
dtype: string
- name: date
dtype: timestamp[s]
- name: file_path
dtype: string
- name: language
dtype: string
- name: language_score
dtype: float64
- name: token_count
dtype: int64
- name: score
dtype: float64
- name: int_score
dtype: int64
splits:
- name: train
num_bytes: 49424526987
num_examples: 8033860
download_size: 27531332308
dataset_size: 49424526987
- config_name: score_2
features:
- name: id
dtype: string
- name: text
dtype: string
- name: safety_metadata
struct:
- name: final_harm_score
dtype: int64
- name: embedding_harm_score
dtype: int64
- name: llm_harm_score
dtype: int64
- name: llm_score_assessment_by_chunk
sequence: string
- name: llm_harm_scores_by_chunk
list:
- name: custom_id
dtype: string
- name: llm_safety_score
dtype: int64
- name: reason
dtype: string
- name: metadata
struct:
- name: dump
dtype: string
- name: url
dtype: string
- name: date
dtype: timestamp[s]
- name: file_path
dtype: string
- name: language
dtype: string
- name: language_score
dtype: float64
- name: token_count
dtype: int64
- name: score
dtype: float64
- name: int_score
dtype: int64
splits:
- name: train
num_bytes: 257652727098
num_examples: 29874316
download_size: 141766458398
dataset_size: 257652727098
- config_name: score_3
features:
- name: id
dtype: string
- name: text
dtype: string
- name: safety_metadata
struct:
- name: final_harm_score
dtype: int64
- name: embedding_harm_score
dtype: int64
- name: llm_harm_score
dtype: int64
- name: llm_score_assessment_by_chunk
sequence: string
- name: llm_harm_scores_by_chunk
list:
- name: custom_id
dtype: string
- name: llm_safety_score
dtype: int64
- name: reason
dtype: string
- name: metadata
struct:
- name: dump
dtype: string
- name: url
dtype: string
- name: date
dtype: timestamp[s]
- name: file_path
dtype: string
- name: language
dtype: string
- name: language_score
dtype: float64
- name: token_count
dtype: int64
- name: score
dtype: float64
- name: int_score
dtype: int64
splits:
- name: train
num_bytes: 35728962911
num_examples: 2704901
download_size: 19906353964
dataset_size: 35728962911
- config_name: score_4
features:
- name: id
dtype: string
- name: text
dtype: string
- name: safety_metadata
struct:
- name: final_harm_score
dtype: int64
- name: embedding_harm_score
dtype: int64
- name: llm_harm_score
dtype: int64
- name: llm_score_assessment_by_chunk
sequence: string
- name: llm_harm_scores_by_chunk
list:
- name: custom_id
dtype: string
- name: llm_safety_score
dtype: int64
- name: reason
dtype: string
- name: metadata
struct:
- name: dump
dtype: string
- name: url
dtype: string
- name: date
dtype: timestamp[s]
- name: file_path
dtype: string
- name: language
dtype: string
- name: language_score
dtype: float64
- name: token_count
dtype: int64
- name: score
dtype: float64
- name: int_score
dtype: int64
splits:
- name: train
num_bytes: 35763743641
num_examples: 2102170
download_size: 20092149434
dataset_size: 35763743641
- config_name: score_5
features:
- name: id
dtype: string
- name: text
dtype: string
- name: safety_metadata
struct:
- name: final_harm_score
dtype: int64
- name: embedding_harm_score
dtype: int64
- name: llm_harm_score
dtype: int64
- name: llm_score_assessment_by_chunk
sequence: string
- name: llm_harm_scores_by_chunk
list:
- name: custom_id
dtype: string
- name: llm_safety_score
dtype: int64
- name: reason
dtype: string
- name: metadata
struct:
- name: dump
dtype: string
- name: url
dtype: string
- name: date
dtype: timestamp[s]
- name: file_path
dtype: string
- name: language
dtype: string
- name: language_score
dtype: float64
- name: token_count
dtype: int64
- name: score
dtype: float64
- name: int_score
dtype: int64
splits:
- name: train
num_bytes: 14872107115
num_examples: 704280
download_size: 8340674635
dataset_size: 14872107115
configs:
- config_name: score_0
data_files:
- split: train
path: score_0/train-*
- config_name: score_1
data_files:
- split: train
path: score_1/train-*
- config_name: score_2
data_files:
- split: train
path: score_2/train-*
- config_name: score_3
data_files:
- split: train
path: score_3/train-*
- config_name: score_4
data_files:
- split: train
path: score_4/train-*
- config_name: score_5
data_files:
- split: train
path: score_5/train-*