Downloads · 30 days
13
3% of all-time downloads
Gptsmall
KovrizhnykhDmitrii/Gptsmall
Gptsmall is a text generation model from KovrizhnykhDmitrii. Use it when you need the model to write or continue text. It is set up for transformers.
Model was trained with sequence length 1024 using transformers lib by SberDevices team on 80B tokens for 3 epochs. After that model was finetuned 1 epoch with sequence length 2048.
All-time downloads
472
Public
Repo size
9.3 GB
Likes
1
Public
Repo makeup
Click a slice to open those files.
.bin3.1 GB · 51%
At a glance
- Task
- Text Generation
- Library
- transformers
- Model type
- gpt2
- Access
- Public
- Created
- Mar 22, 2023
- Updated
- Mar 22, 2023
- SHA
- 73c7e5f9
Model card
From the Hugging Face model README
rugpt3large_based_on_gpt2
Model was trained with sequence length 1024 using transformers lib by SberDevices team on 80B tokens for 3 epochs. After that model was finetuned 1 epoch with sequence length 2048.
Total training time was around 14 days on 128 GPUs for 1024 context and few days on 16 GPUs for 2048 context.
Final perplexity on test set is 13.6.
- Task
- Text Generation
- Library
- transformers
- Type
- gpt2
- Languages
- ru
- Created
- Mar 22, 2023
- Updated
- Mar 22, 2023