The dataset viewer is not available because its heuristics could not detect any supported data files. You can try uploading some data files, or configuring the data files location manually.
More than one training run goes into making a large language model, but developers rarely release the small models and datasets they experiment with during the development process. How do they decide what dataset to use for pretraining or which benchmarks to hill climb on? To empower open exploration of these questions, we release DataDecide—a suite of models we pretrain on 25 corpora with differing sources, deduplication, and filtering up to 100B tokens, over 14 different model sizes ranging from 4M parameters up to 1B parameters (more than 30k model checkpoints in total).
We call the 25 corpora we train on data recipes as they range across popular corpora including Dolma, DCLM, RefinedWeb, C4, and FineWeb as well as combinations of interventions on these datasets such as source mixing, deduplication, and filtering. This HuggingFace Dataset contains the tokenized data used to build these recipes with OLMo. Uploading is in progress.
| Source | Recipe | Description |
|---|---|---|
| Dolma1.7 | Original | The 1.7 release of the Dolma dataset (Soldaini et al., 2024), a 2.3 trillion token corpus sampling sources commonly used in LM training. |
| Dolma1.7 | No code | Dolma1.7 with code-related subsets (Starcoder, StackExchange) removed. |
| Dolma1.7 | No math, code | Dolma1.7 excluding OpenWebMath, arXiv STEM papers, Starcoder, StackExchange, and Algebraic Stack. |
| Dolma1.7 | No Reddit | Dolma1.7 with Reddit subset excluded. |
| Dolma1.7 | No Flan | Dolma1.7 with Flan subset removed. |
| Dolma1.6++ | Original | Dolma1.6 with additional sources from Dolma1.7: RedPajama ArXiv, OpenWebMath, Algebraic Stack, Flan, Starcoder, and Falcon. |
| C4 | Original | The C4 dataset (Raffel et al., 2020) as processed in Dolma1.7, derived from April 2019 Common Crawl with automatic filtering. |
| FineWeb-Pro | Original | FineWeb-Pro (Zhou et al., 2024), created using a model-guided approach to apply programmatic cleaning over FineWeb. |
| FineWeb-Edu | Original | FineWeb-Edu (Benallal et al., 2024), deduplicated subset of SmolLM-Corpus filtered by an educational quality classifier. |
| Falcon | Original | Falcon RefinedWeb (Penedo et al., 2023) as used in Dolma1.7, built from all Common Crawl through June 2023, aggressively filtered. |
| Falcon+CC | Original | Unfiltered combination of Falcon RefinedWeb and Dolma1.7's Common Crawl data. |
| Falcon+CC | QC 10% | Top 10% of Falcon+CC by reproduction of the DCLM quality filter (Li et al., 2024). |
| Falcon+CC | QC 20% | Top 20% of Falcon+CC by reproduced DCLM filter. |
| Falcon+CC | QC Orig 10% | Top 10% using the original DCLM-provided quality filter. |
| Falcon+CC | QC Tulu 10% | Top 10% filtered using a classifier trained on pre-release Tulu-v3 (Lambert et al., 2024). |
| DCLM-Baseline | Original | DCLM-Baseline from Li et al. (2024). |
| DCLM-Baseline | QC 7% FW2 | Top 7% by DCLM filter, filtered with FineWeb-Edu, keeping only documents scored ≥2. |
| DCLM-Baseline | QC 7% FW3 | Same as above but restricted to documents scored ≥3. |
| DCLM-Baseline | QC FW 10% | Filtered using FineWeb-Edu classifier, top 10% retained. |
| DCLM-Baseline | QC FW 3% | Same as above but only top 3% retained. |
| DCLM-Baseline | QC 10% | 10% retained using classifier fine-tuned on OpenHermes and Reddit ELI5. |
| DCLM-Baseline | QC 20% | Same as above, but retaining top 20%. |
| DCLM-Baseline 25% / Dolma 75% | 75% Dolma / 25% DCLM | Mixed dataset: 75% Dolma1.7 and 25% DCLM-Baseline. |
| DCLM-Baseline 50% / Dolma 50% | 50% Dolma / 50% DCLM | Mixed dataset: 50% Dolma1.7 and 50% DCLM-Baseline. |
| DCLM-Baseline 75% / Dolma 25% | 25% Dolma / 75% DCLM | Mixed dataset: 25% Dolma1.7 and 75% DCLM-Baseline. |
For each of our 25 datasets and 14 model sizes, we train a model linked below. Each has intermediate checkpoints (uploading after initial release), runs over 3 random seeds. All models finish training at a token to parameter ratio of 100 (e.g., 1B parameters -> 100B tokens).
BibTeX:
@article{MagnussonDataDecide2025,
title={{DataDecide: How to Predict Best Pretraining Data with Small Experiments}},
author={Ian Magnusson and Nguyen Tai and Ben Bogin and David Heineman and Jena Hwang and Luca Soldaini and Akshita Bhagia and Jiacheng Liu and Dirk Groeneveld and Oyvind Tafjord and Noah A. Smith and Pang Wei Koh and Jesse Dodge},
year={2025},
journal={arXiv preprint},
}