metadata
dataset_info:
features:
- name: text
dtype: string
- name: sentiment
dtype: int64
- name: respect
dtype: int64
- name: insult
dtype: int64
- name: humiliate
dtype: int64
- name: status
dtype: int64
- name: dehumanize
dtype: int64
- name: violence
dtype: int64
- name: genocide
dtype: int64
- name: attack_defend
dtype: int64
- name: hatespeech
dtype: int64
splits:
- name: train
num_bytes: 20985424
num_examples: 95433
- name: test
num_bytes: 2621502
num_examples: 11929
- name: validation
num_bytes: 2621368
num_examples: 11930
download_size: 9072984
dataset_size: 26228294
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*
Simplified version of Measuring Hate Speech using our custom class thresholds. Original dataset: https://huggingface.co/datasets/ucberkeley-dlab/measuring-hate-speech