Stack_TokLlama / README.md
Trevor Dohm
Tokenized bro
d1a261c verified
---
dataset_info:
features:
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
splits:
- name: train
num_bytes: 7827160
num_examples: 1100
download_size: 1628463
dataset_size: 7827160
configs:
- config_name: default
data_files:
- split: train
path: data/bro/train-*
---