metadata
dataset_info:
features:
- name: name
dtype: string
- name: body
dtype: string
- name: score
dtype: int64
- name: controversiality
dtype: int64
- name: created
dtype: timestamp[us]
- name: author
dtype: string
- name: collapsed
dtype: bool
- name: edited
dtype: timestamp[us]
- name: gilded
dtype: int64
- name: id
dtype: string
- name: locked
dtype: bool
- name: permalink
dtype: string
- name: stickied
dtype: bool
- name: ups
dtype: int64
splits:
- name: train
num_bytes: 613095589
num_examples: 1508942
download_size: 291543991
dataset_size: 613095589
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
license: odc-by
task_categories:
- text-generation
- feature-extraction
- text-classification
language:
- en
tags:
- reddit
LocalLLaMA-comments
A companion dataset to pszemraj/LocalLLaMA-posts. Time frame is in sync (up through Tue Mar 3 9PM EST 2026)