Datasets:

Modalities:
Tabular
Text
Formats:
parquet
Languages:
English
ArXiv:
DOI:
Libraries:
Datasets
Dask
License:
loubnabnl HF staff commited on
Commit
8cb2cd5
β€’
1 Parent(s): ede1fa7

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -3
README.md CHANGED
@@ -500,12 +500,11 @@ The classifier is available at: [https://huggingface.co/HuggingFaceFW/fineweb-ed
500
 
501
  We investigated the impact of using different thresholds for the filtering and found that threshold 3 gave the best overall results. Although using a threshold higher than 3 improves performance on knowledge and reasoning intensive benchmarks, it significantly degrades performance on HellaSwag and PIQA.
502
 
503
- We then built πŸ“š FineWeb-Edu by filtering out samples with scores lower than 3. This removed 92% of the dataset, leaving us with 1.3T educational tokens. Our ablation demonstrated that this refined dataset surpasses 🍷 FineWeb and all other open web datasets, with remarkable improvements on educational benchmarks such as MMLU, ARC, and OpenBookQA. To retain more tokens, we also experimented with a less strict threshold of 2 instead of 3. While being less performant than using threshold 3, it still outperformed FineWeb and it preserved 5.4T tokens.
504
- The plot below compare FineWeb-Edu to other web datasets:
505
 
506
  ![image/png](https://cdn-uploads.huggingface.co/production/uploads/61c141342aac764ce1654e43/hJlyTgDzZpYuxO9LUm0PF.png)
507
 
508
- We release these two dataset as [FineWeb-Edu](https://huggingface.co/datasets/HuggingFaceFW/fineweb-edu) and [FineWeb-Edu-score-2](https://huggingface.co/datasets/HuggingFaceFW/fineweb-edu-score-2) along with the [classifier](https://huggingface.co/HuggingFaceFW/fineweb-edu-classifier).
509
 
510
  You will find all the ablation models in [this collection](https://huggingface.co/collections/HuggingFaceFW/ablation-models-662457b0d213e8c14fe47f32). The FineWeb-Edu ablation model (trained on 350B tokens) is available at [https://huggingface.co/HuggingFaceFW/ablation-model-fineweb-edu](https://huggingface.co/HuggingFaceFW/ablation-model-fineweb-edu).
511
 
 
500
 
501
  We investigated the impact of using different thresholds for the filtering and found that threshold 3 gave the best overall results. Although using a threshold higher than 3 improves performance on knowledge and reasoning intensive benchmarks, it significantly degrades performance on HellaSwag and PIQA.
502
 
503
+ We then built πŸ“š FineWeb-Edu by filtering out samples with scores lower than 3. This removed 92% of the dataset, leaving us with 1.3T educational tokens. Our ablation demonstrated that this refined dataset surpasses 🍷 FineWeb and all other open web datasets, with remarkable improvements on educational benchmarks such as MMLU, ARC, and OpenBookQA. The plot below compares FineWeb-Edu to other web datasets:
 
504
 
505
  ![image/png](https://cdn-uploads.huggingface.co/production/uploads/61c141342aac764ce1654e43/hJlyTgDzZpYuxO9LUm0PF.png)
506
 
507
+ To retain more tokens, we also experimented with a less strict threshold of 2 instead of 3. While being less performant than using threshold 3, it still outperformed FineWeb and it preserved 5.4T tokens. We release these two dataset as [FineWeb-Edu](https://huggingface.co/datasets/HuggingFaceFW/fineweb-edu) and [FineWeb-Edu-score-2](https://huggingface.co/datasets/HuggingFaceFW/fineweb-edu-score-2) along with the [classifier](https://huggingface.co/HuggingFaceFW/fineweb-edu-classifier).
508
 
509
  You will find all the ablation models in [this collection](https://huggingface.co/collections/HuggingFaceFW/ablation-models-662457b0d213e8c14fe47f32). The FineWeb-Edu ablation model (trained on 350B tokens) is available at [https://huggingface.co/HuggingFaceFW/ablation-model-fineweb-edu](https://huggingface.co/HuggingFaceFW/ablation-model-fineweb-edu).
510