Downloads · 30 days
124
46% of all-time downloads
GPT2_fine_tuned_SP_GPT2_config_ESM_tokenizer_50kTrembl_30epochs
Jacopo-gab/GPT2_fine_tuned_SP_GPT2_config_ESM_tokenizer_50kTrembl_30epochs
GPT2_fine_tuned_SP_GPT2_config_ESM_tokenizer_50kTrembl_30epochs is a text generation model from Jacopo-gab. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as mit.
should probably proofread and complete it, then remove this comment. --
All-time downloads
270
Public
Repo size
687 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin344 MB · 100%
At a glance
- Task
- Text Generation
- Library
- transformers
- License
- mit
- Model type
- gpt2
- Access
- Public
- Created
- Jun 11, 2023
- Updated
- Jun 11, 2023
- SHA
- cc1380a0
Try a prompt
Model card
From the Hugging Face model README
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
GPT2_fine_tuned_SP_GPT2_config_ESM_tokenizer_50kTrembl_30epochs
This model is a fine-tuned version of gpt2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.3935
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.001
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 3981
- num_epochs: 30
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 3.2753 | 1.0 | 155 | 2.6914 |
| 2.651 | 2.0 | 311 | 2.6160 |
| 2.6217 | 3.0 | 466 | 2.5850 |
| 2.5527 | 4.0 | 622 | 2.4774 |
| 2.4127 | 5.0 | 777 | 2.1669 |
| 2.0779 | 6.0 | 933 | 1.8001 |
| 1.7924 | 7.0 | 1088 | 1.5899 |
| 1.5674 | 8.0 | 1244 | 1.4479 |
| 1.4213 | 9.0 | 1399 | 1.3477 |
| 1.2901 | 10.0 | 1555 | 1.2820 |
| 1.199 | 11.0 | 1710 | 1.2308 |
| 1.1071 | 12.0 | 1866 | 1.1902 |
| 1.0392 | 13.0 | 2021 | 1.1722 |
| 0.9683 | 14.0 | 2177 | 1.1432 |
| 0.9162 | 15.0 | 2332 | 1.1426 |
| 0.8584 | 16.0 | 2488 | 1.1227 |
| 0.8148 | 17.0 | 2643 | 1.1374 |
| 0.7638 | 18.0 | 2799 | 1.1213 |
| 0.7278 | 19.0 | 2954 | 1.1390 |
| 0.6844 | 20.0 | 3110 | 1.1592 |
| 0.6554 | 21.0 | 3265 | 1.1621 |
| 0.6214 | 22.0 | 3421 | 1.1772 |
| 0.5978 | 23.0 | 3576 | 1.1828 |
| 0.5706 | 24.0 | 3732 | 1.2075 |
| 0.5538 | 25.0 | 3887 | 1.2180 |
| 0.5329 | 26.0 | 4043 | 1.2230 |
| 0.4948 | 27.0 | 4198 | 1.2660 |
| 0.4087 | 28.0 | 4354 | 1.3091 |
| 0.3373 | 29.0 | 4509 | 1.3572 |
| 0.2973 | 29.9 | 4650 | 1.3935 |
Framework versions
- Transformers 4.30.1
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
- Task
- Text Generation
- Library
- transformers
- Type
- gpt2
- License
- mit
- Created
- Jun 11, 2023
- Updated
- Jun 11, 2023