Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- sv
|
4 |
+
|
5 |
+
---
|
6 |
+
|
7 |
+
# Megatron-BERT-base Swedish 125k
|
8 |
+
|
9 |
+
This BERT model was trained using the Megatron-LM library.
|
10 |
+
The size of the model is a regular BERT-base with 110M parameters.
|
11 |
+
The model was trained on about 70GB of data, consisting mostly of OSCAR and Swedish newspaper text curated by the National Library of Sweden.
|
12 |
+
|
13 |
+
Training was done for 125k training steps. Its [sister model](https://huggingface.co/KBLab/megatron-bert-base-swedish-cased-600k) used the same setup, but was instead trained for 600k steps.
|
14 |
+
|
15 |
+
|
16 |
+
The model has three sister models trained on the same dataset:
|
17 |
+
- [🤗 BERT Swedish](https://huggingface.co/KBLab/bert-base-swedish-cased-new)
|
18 |
+
- [Megatron-BERT-base-600k](https://huggingface.co/KBLab/megatron-bert-base-swedish-cased-600k)
|
19 |
+
- [Megatron-BERT-large-110k]()
|