Datasets:

HuggingFaceFW
/

fineweb-edu

Tasks:

Text Generation

Modalities:

Formats:

Languages:

Size:

ArXiv:

DOI:

Libraries:

License:

Dataset card Viewer Files Files and versions Community

loubnabnl HF staff commited on Jun 2

Commit

ab8f77d

•

1 Parent(s): 75fa67c

remove todo

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -499,7 +499,7 @@ We fine-tuned a Bert-like regression model using these annotations, based on [Sn
 The classifier is available at: [https://huggingface.co/HuggingFaceFW/fineweb-edu-classifier/](HuggingFaceFW/fineweb-edu-classifier/)
 ### Filtering and results
-**Note**: You can find more details about the ablations and results in the FineWeb blog post (TODO).
 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.

 The classifier is available at: [https://huggingface.co/HuggingFaceFW/fineweb-edu-classifier/](HuggingFaceFW/fineweb-edu-classifier/)
 ### Filtering and results
+**Note**: You can find more details about the ablations and results in the FineWeb [blog post](https://huggingface.co/spaces/HuggingFaceFW/blogpost-fineweb-v1).
 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.