Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,60 @@
|
|
1 |
---
|
2 |
-
license:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
license: cc-by-4.0
|
3 |
+
datasets:
|
4 |
+
- clarin-knext/msmarco-pl
|
5 |
+
- clarin-knext/nq-pl
|
6 |
+
- clarin-knext/hotpotqa-pl
|
7 |
+
- clarin-knext/scidocs-pl
|
8 |
+
- clarin-knext/nfcorpus-pl
|
9 |
+
- clarin-knext/dbpedia-pl
|
10 |
+
- clarin-knext/trec-covid-pl
|
11 |
+
- clarin-knext/quora-pl
|
12 |
+
- clarin-knext/arguana-pl
|
13 |
+
- clarin-knext/fiqa-pl
|
14 |
+
- radlab/wikipedia-pl
|
15 |
+
- radlab/legal-mc4-pl
|
16 |
+
language:
|
17 |
+
- pl
|
18 |
+
library_name: transformers
|
19 |
+
tags:
|
20 |
+
- gpt2
|
21 |
+
- from-scratch
|
22 |
+
- polish-gpt2
|
23 |
---
|
24 |
+
|
25 |
+
## Description
|
26 |
+
This is the polish gpt2 model in small architecture.
|
27 |
+
This model was released on 30.11.2023 and it is the newest version of `radlab/polish-gpt2-small`
|
28 |
+
|
29 |
+
|
30 |
+
## Datasets
|
31 |
+
Data which are used to train this model:
|
32 |
+
- clarin-knext/msmarco-pl
|
33 |
+
- clarin-knext/nq-pl
|
34 |
+
- clarin-knext/hotpotqa-pl
|
35 |
+
- clarin-knext/scidocs-pl
|
36 |
+
- clarin-knext/nfcorpus-pl
|
37 |
+
- clarin-knext/dbpedia-pl
|
38 |
+
- clarin-knext/trec-covid-pl
|
39 |
+
- clarin-knext/quora-pl
|
40 |
+
- clarin-knext/arguana-pl
|
41 |
+
- clarin-knext/fiqa-pl
|
42 |
+
- radlab/wikipedia-pl
|
43 |
+
- radlab/legal-mc4-pl
|
44 |
+
- own corpora not published yet
|
45 |
+
|
46 |
+
It is about 30,5 GB of data which is 3 times more than the prevoius version.
|
47 |
+
|
48 |
+
|
49 |
+
## Metrics from W&B
|
50 |
+
|
51 |
+
|
52 |
+
![image/png](https://cdn-uploads.huggingface.co/production/uploads/644addfe9279988e0cbc296b/BBhEM3RL4rM1rOwTG-1RC.png)
|
53 |
+
|
54 |
+
|
55 |
+
![image/png](https://cdn-uploads.huggingface.co/production/uploads/644addfe9279988e0cbc296b/Z7BKY2vKPXWkQZEybt2rW.png)
|
56 |
+
|
57 |
+
![image/png](https://cdn-uploads.huggingface.co/production/uploads/644addfe9279988e0cbc296b/upsovWy0lVohPGjR1AGmN.png)
|
58 |
+
|
59 |
+
## Changelog
|
60 |
+
- _2023.11.30_ - new dataset
|