---
dataset_info:
  features:
  - name: input_ids
    sequence: int32
  splits:
  - name: train
    num_bytes: 5836120400.0
    num_examples: 1423444
  download_size: 2517314986
  dataset_size: 5836120400.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---

Pre-tokenized dataset of the first 1 million lines of [monology/pile-uncopyrighted](https://huggingface.co/datasets/monology/pile-uncopyrighted), tokenized for Gemma-2 using SAELens.

This dataset has 1024 context size and about 1.5B tokens.