Andrey Kutuzov commited on
Commit
87def6d
1 Parent(s): a1d3a5d
Files changed (6) hide show
  1. README.md +27 -3
  2. config.json +17 -0
  3. pytorch_model.bin +3 -0
  4. tf_model.h5 +3 -0
  5. tokenizer_config.json +3 -0
  6. vocab.txt +0 -0
README.md CHANGED
@@ -1,3 +1,27 @@
1
- ---
2
- license: cc-by-4.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: no
3
+ license: cc-by-4.0
4
+ pipeline_tag: fill-mask
5
+ tags:
6
+ - norwegian
7
+ - bert
8
+ thumbnail: https://raw.githubusercontent.com/ltgoslo/NorBERT/main/Norbert.png
9
+ ---
10
+ ## Quickstart
11
+ **Release 2.0** (February 7, 2022)
12
+ Trained on the very large corpus of Norwegian (C4 + NCC, about 15 billion word tokens).
13
+ Features a 50 000 words vocabulary and was trained using Whole Word Masking.
14
+
15
+ Download the model here:
16
+ * Cased Norwegian BERT Base 2.0 (NorBERT 2): [221.zip](http://vectors.nlpl.eu/repository/20/221.zip)
17
+ More about NorBERT training corpora, training procedure and evaluation benchmarks: http://norlm.nlpl.eu/
18
+
19
+ Associated code: https://github.com/ltgoslo/NorBERT
20
+
21
+ Check this paper for more details:
22
+ _Andrey Kutuzov, Jeremy Barnes, Erik Velldal, Lilja Øvrelid, Stephan Oepen. [Large-Scale Contextualised Language Modelling for Norwegian](https://arxiv.org/abs/2104.06546), NoDaLiDa'21 (2021)_
23
+
24
+ NorBERT was trained as a part of NorLM, a joint initiative of the projects [EOSC-Nordic](https://www.eosc-nordic.eu/) (European Open Science Cloud),
25
+ coordinated by the [Language Technology Group](https://www.mn.uio.no/ifi/english/research/groups/ltg/) (LTG) at the University of Oslo.
26
+
27
+ The computations were performed on resources provided by UNINETT Sigma2 - the National Infrastructure for High Performance Computing and Data Storage in Norway.
config.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "BertForMaskedLM"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "hidden_act": "gelu",
7
+ "hidden_dropout_prob": 0.1,
8
+ "hidden_size": 768,
9
+ "initializer_range": 0.02,
10
+ "intermediate_size": 3072,
11
+ "max_position_embeddings": 512,
12
+ "model_type": "bert",
13
+ "num_attention_heads": 12,
14
+ "num_hidden_layers": 12,
15
+ "type_vocab_size": 2,
16
+ "vocab_size": 50104
17
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3c15dc6df9ed0cd6db1726ea7a2d5e8a7e73fa950065cb60436f09c1e7a6609c
3
+ size 500746539
tf_model.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8741bc2e9666330a411b6e2d1d59280e31c047d141b73f82976ee3d2445b7449
3
+ size 498346776
tokenizer_config.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "do_lower_case": false
3
+ }
vocab.txt ADDED
The diff for this file is too large to render. See raw diff