Update README.md
Browse files
README.md
CHANGED
|
@@ -9,6 +9,13 @@ language:
|
|
| 9 |
CSMPT7b is a large Czech language model continously pretrained on 272b training tokens from English [MPT7b](https://huggingface.co/mosaicml/mpt-7b) model. Model was pretrained on ~67b token [Large Czech Collection](https://huggingface.co/datasets/BUT-FIT/but_lcc) using Czech tokenizer, obtained using our vocabulary swap method (see below).
|
| 10 |
Training was done on [Karolina](https://www.it4i.cz/en) cluster.
|
| 11 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 12 |
# Evaluation
|
| 13 |
Dev eval at CS-HellaSwag (automatically translated HellaSwag benchmark).
|
| 14 |
| Model | CS-HellaSwag Accuracy |
|
|
|
|
| 9 |
CSMPT7b is a large Czech language model continously pretrained on 272b training tokens from English [MPT7b](https://huggingface.co/mosaicml/mpt-7b) model. Model was pretrained on ~67b token [Large Czech Collection](https://huggingface.co/datasets/BUT-FIT/but_lcc) using Czech tokenizer, obtained using our vocabulary swap method (see below).
|
| 10 |
Training was done on [Karolina](https://www.it4i.cz/en) cluster.
|
| 11 |
|
| 12 |
+
|
| 13 |
+
# <span style="color:red">BUT Model Roster</span>
|
| 14 |
+
- [BUT-FIT/CSTinyLlama-1.2B](https://huggingface.co/BUT-FIT/CSTinyLlama-1.2B)
|
| 15 |
+
- [BUT-FIT/Czech-GPT-2-XL-133k](https://huggingface.co/BUT-FIT/Czech-GPT-2-XL-133k)
|
| 16 |
+
- [BUT-FIT/csmpt7b](https://huggingface.co/BUT-FIT/csmpt7b)
|
| 17 |
+
|
| 18 |
+
|
| 19 |
# Evaluation
|
| 20 |
Dev eval at CS-HellaSwag (automatically translated HellaSwag benchmark).
|
| 21 |
| Model | CS-HellaSwag Accuracy |
|