File size: 469 Bytes
d6f784b
 
 
 
 
 
bbdb033
d6f784b
 
ec2adc0
 
 
9774b72
 
f8edff3
 
9774b72
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
---
language:
- lu
tags:
- text
- MLM
license: mit
---

## BERT Medium for Luxembourgish

Created from a dataset with 1M Luxembourgish sentences from Wikipedia. Corpus has approx. 16M words.

The MLM objective was trained. The BERT model has parameters `L=8` and `H=512`. Vocabulary has 70K word pieces.

Final loss scores, after 3 epochs:

- Final train loss: 4.230
- Final train perplexity: 68.726
- Final validation loss: 4.074
- Final validation perplexity: 58.765