metadata
license: apache-2.0
task_categories:
- text-generation
language:
- en
configs:
- config_name: default
data_files:
- split: train
path:
- shard_00000.parquet
Dataset
Mixed pretraining corpus - 100,000 rows, 1 shards.
Deduped (exact=False, fuzzy=False).