Update README.md

README.md

@@ -31,7 +31,7 @@ alephbert.eval()
 
 ## Training data
 1. OSCAR [(Ortiz, 2019)](https://oscar-corpus.com/) Hebrew section (10GB text, 20M sentences).
-2. Hebrew dump of [Wikipedia](https://dumps.wikimedia.org/hewiki/latest/) (650 MB text, 3.8M sentences).
+2. Hebrew dump of [Wikipedia](https://dumps.wikimedia.org/hewiki/latest/) (650 MB text, 3M sentences).
 3. Hebrew Tweets collected from the Twitter sample stream (7G text, 70M sentences).
 
 ## Training procedure
@@ -43,7 +43,7 @@ To optimize training time we split the data into 4 sections based on max number
 1. num tokens < 32 (70M sentences)
 2. 32 <= num tokens < 64 (12M sentences)
 3. 64 <= num tokens < 128 (10M sentences)
-4. 128 <= num tokens < 512 (70M sentences)
+4. 128 <= num tokens < 512 (1.5M sentences)
 
 Each section was first trained for 5 epochs with an initial learning rate set to 1e-4. Then each section was trained for another 5 epochs with an initial learning rate set to 1e-5, for a total of 10 epochs.