vmkhlv commited on
Commit
4482c35
1 Parent(s): 8201db0

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +18 -4
README.md CHANGED
@@ -8,11 +8,25 @@ thumbnail: "https://github.com/sberbank-ai/ru-gpts"
8
  ---
9
 
10
  # rugpt3large\_based\_on\_gpt2
11
- Model was trained with sequence length 1024 using transformers lib by [SberDevices](https://sberdevices.ru/) team on 80B tokens for 3 epochs. After that model was finetuned 1 epoch with sequence length 2048.
12
 
13
- Total training time was around 14 days on 128 GPUs for 1024 context and few days on 16 GPUs for 2048 context.
14
- Final perplexity on test set is `13.6`.
 
 
15
 
16
  # Authors
17
  + NLP core team RnD [Telegram channel](https://t.me/nlpcoreteam):
18
- + Dmitry Zmitrovich
 
 
 
 
 
 
 
 
 
 
 
 
 
8
  ---
9
 
10
  # rugpt3large\_based\_on\_gpt2
11
+ The model architecture design, pretraining, and evaluation are documented in our preprint: [**A Family of Pretrained Transformer Language Models for Russian**](https://arxiv.org/abs/2309.10931).
12
 
13
+ The model was trained with sequence length 1024 using transformers lib by the [SberDevices](https://sberdevices.ru/) team on 80B tokens for 3 epochs. After that, the model was finetuned 1 epoch with sequence length 2048.
14
+
15
+ Total training time was around 14 days on 128 GPUs for 1024 context and a few days on 16 GPUs for 2048 context.
16
+ The final perplexity on the test set is `13.6`.
17
 
18
  # Authors
19
  + NLP core team RnD [Telegram channel](https://t.me/nlpcoreteam):
20
+ + Dmitry Zmitrovich
21
+
22
+ # Cite us
23
+ ```
24
+ @misc{zmitrovich2023family,
25
+ title={A Family of Pretrained Transformer Language Models for Russian},
26
+ author={Dmitry Zmitrovich and Alexander Abramov and Andrey Kalmykov and Maria Tikhonova and Ekaterina Taktasheva and Danil Astafurov and Mark Baushenko and Artem Snegirev and Tatiana Shavrina and Sergey Markov and Vladislav Mikhailov and Alena Fenogenova},
27
+ year={2023},
28
+ eprint={2309.10931},
29
+ archivePrefix={arXiv},
30
+ primaryClass={cs.CL}
31
+ }
32
+ ```