Hugging Face's logo

Datasets:
allenai
/
olmo-mix-1124

Tasks:
Text Generation
Modalities:
Text
Languages:
English
Size:
100M - 1B
License:
Dataset card Data Studio Files
xet
Community
15
olmo-mix-1124 / README.md
baileyk's picture
baileyk
fix typo in `features` (#13)
8162bd7 verified
preview code
|
raw
history blame contribute delete
5.22 kB
metadata
license: odc-by
task_categories:
  - text-generation
language:
  - en
pretty_name: OLMo 2 Mix (November 2024)
size_categories:
  - 1B<n<10B
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/**/*
    features:
      - name: text
        dtype: string
      - name: added
        dtype: string
      - name: created
        dtype: string
      - name: attributes
        dtype: string
      - name: doc
        dtype: string
      - name: id
        dtype: string
      - name: metadata
        dtype: string
      - name: source
        dtype: string
      - name: version
        dtype: string
  - config_name: algebraic-stack
    data_files:
      - split: train
        path: data/algebraic-stack/**/*
    features:
      - name: text
        dtype: string
      - name: added
        dtype: string
      - name: created
        dtype: string
      - name: attributes
        dtype: string
      - name: doc
        dtype: string
      - name: id
        dtype: string
      - name: metadata
        dtype: string
      - name: source
        dtype: string
      - name: version
        dtype: string
  - config_name: arxiv
    data_files:
      - split: train
        path: data/arxiv/**/*
    features:
      - name: text
        dtype: string
      - name: added
        dtype: string
      - name: created
        dtype: string
      - name: attributes
        dtype: string
      - name: doc
        dtype: string
      - name: id
        dtype: string
      - name: metadata
        dtype: string
      - name: source
        dtype: string
      - name: version
        dtype: string
  - config_name: dclm
    data_files:
      - split: train
        path: data/dclm/**/*
    features:
      - name: text
        dtype: string
      - name: added
        dtype: string
      - name: created
        dtype: string
      - name: attributes
        dtype: string
      - name: doc
        dtype: string
      - name: id
        dtype: string
      - name: metadata
        dtype: string
      - name: source
        dtype: string
      - name: version
        dtype: string
  - config_name: open-web-math
    data_files:
      - split: train
        path: data/open-web-math/**/*
    features:
      - name: text
        dtype: string
      - name: added
        dtype: string
      - name: created
        dtype: string
      - name: attributes
        dtype: string
      - name: doc
        dtype: string
      - name: id
        dtype: string
      - name: metadata
        dtype: string
      - name: source
        dtype: string
      - name: version
        dtype: string
  - config_name: pes2o
    data_files:
      - split: train
        path: data/pes2o/*
    features:
      - name: text
        dtype: string
      - name: added
        dtype: string
      - name: created
        dtype: string
      - name: attributes
        dtype: string
      - name: doc
        dtype: string
      - name: id
        dtype: string
      - name: metadata
        dtype: string
      - name: source
        dtype: string
      - name: version
        dtype: string
  - config_name: starcoder
    data_files:
      - split: train
        path: data/starcoder/**/*
    features:
      - name: text
        dtype: string
      - name: added
        dtype: string
      - name: created
        dtype: string
      - name: attributes
        dtype: string
      - name: doc
        dtype: string
      - name: id
        dtype: string
      - name: metadata
        dtype: string
      - name: source
        dtype: string
      - name: version
        dtype: string
  - config_name: wiki
    data_files:
      - split: train
        path: data/wiki/*
    features:
      - name: text
        dtype: string
      - name: added
        dtype: string
      - name: created
        dtype: string
      - name: attributes
        dtype: string
      - name: doc
        dtype: string
      - name: id
        dtype: string
      - name: metadata
        dtype: string
      - name: source
        dtype: string
      - name: version
        dtype: string

OLMo 2 (November 2024) Pretraining set

Collection of data used to train OLMo-2-1124 models. The majority of this dataset comes from DCLM-Baseline with no additional filtering, but we provide the explicit breakdowns below.

Name Tokens Bytes (uncompressed) Documents License
DCLM-Baseline 3.70T 21.3TB 2.95B CC-BY-4.0
Arxiv 20.8B 77.2GB 3.95M ODC-BY
pes2o 58.6B 412GB 38M ODC-BY
starcoder 83.0B 458GB 78.7M ODC-BY
Algebraic-stack 11.8B 44.0GB 2.83M ODC-BY
OpenWebMath 12.2B 47.23GB 2.89M ODC-BY
Wiki 3.66B 18.1GB 6.17M ODC-BY
Total 3.90T 22.4TB 3.08B ODC-BY

Please refer to the OLMo2 Tech Report for further details.

Licensing Information

This collection is released under the Open Data Commons Attribution License (ODC-By) v1.0 license. The use of this dataset is also subject to CommonCrawl's Terms of Use.

Citation

A technical manuscript is forthcoming!