Hugging Face's logo

Datasets:
lucadiliello
/
wikipedia_512_pretraining

Languages:
English
Size Categories:
1M<n<10M
Tags:
Dataset card Viewer Files Community
lucadiliello commited on
Commit
dabe901
•
1 Parent(s): e42e968

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -23,4 +23,4 @@ size_categories:
23
  ---
24
  # Dataset Card for "wikipedia_512_pretraining"
25
 
26
- Wikipedia preprocessed for pretraining of models. Each sample in the dataset has an average tokenized length of 512 tokens.
 
23
  ---
24
  # Dataset Card for "wikipedia_512_pretraining"
25
 
26
+ Wikipedia preprocessed for pretraining of models. Each sample in the dataset has an average tokenized length of 512 `RoBERTa-Base` tokens.