Downloads · 30 days
12
21% of all-time downloads
sourceformer-epoch30
eerichmond33/sourceformer-epoch30
sourceformer-epoch30 is a text generation model from eerichmond33. 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
58
Public
Repo size
5.5 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin2.7 GB · 100%
At a glance
- Task
- Text Generation
- Library
- transformers
- License
- mit
- Model type
- gpt_neo
- Access
- Public
- Created
- Jul 19, 2023
- Updated
- Jul 19, 2023
- SHA
- c52d1c18
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. -->
v29
This model is a fine-tuned version of EleutherAI/gpt-neo-1.3B on the None dataset. It achieves the following results on the evaluation set:
- Loss: 5.1914
- Accuracy: 0.3454
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: 1e-05
- train_batch_size: 4
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- total_train_batch_size: 8
- total_eval_batch_size: 2
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 210
- num_epochs: 30.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.7512 | 1.0 | 72 | 2.7871 | 0.4451 |
| 2.6444 | 2.0 | 144 | 2.7832 | 0.4452 |
| 2.5407 | 3.0 | 216 | 2.7930 | 0.4429 |
| 2.4257 | 4.0 | 288 | 2.8125 | 0.4411 |
| 2.3014 | 5.0 | 360 | 2.8477 | 0.4377 |
| 2.1749 | 6.0 | 432 | 2.8945 | 0.4328 |
| 2.0323 | 7.0 | 504 | 2.9668 | 0.4256 |
| 1.8754 | 8.0 | 576 | 3.0352 | 0.4197 |
| 1.7043 | 9.0 | 648 | 3.1504 | 0.4117 |
| 1.5224 | 10.0 | 720 | 3.2773 | 0.4029 |
| 1.3273 | 11.0 | 792 | 3.4141 | 0.3945 |
| 1.1364 | 12.0 | 864 | 3.5625 | 0.3869 |
| 0.9499 | 13.0 | 936 | 3.7422 | 0.3784 |
| 0.7793 | 14.0 | 1008 | 3.8906 | 0.3731 |
| 0.6176 | 15.0 | 1080 | 4.0625 | 0.3669 |
| 0.4873 | 16.0 | 1152 | 4.1836 | 0.3630 |
| 0.3717 | 17.0 | 1224 | 4.3281 | 0.3589 |
| 0.2797 | 18.0 | 1296 | 4.4570 | 0.3564 |
| 0.2054 | 19.0 | 1368 | 4.5703 | 0.3539 |
| 0.1517 | 20.0 | 1440 | 4.6719 | 0.3525 |
| 0.1115 | 21.0 | 1512 | 4.7539 | 0.3511 |
| 0.0854 | 22.0 | 1584 | 4.8359 | 0.3494 |
| 0.0669 | 23.0 | 1656 | 4.9062 | 0.3492 |
| 0.0494 | 24.0 | 1728 | 4.9609 | 0.3486 |
| 0.0377 | 25.0 | 1800 | 5.0273 | 0.3472 |
| 0.0302 | 26.0 | 1872 | 5.0625 | 0.3471 |
| 0.026 | 27.0 | 1944 | 5.1406 | 0.3455 |
| 0.0255 | 28.0 | 2016 | 5.1211 | 0.3465 |
| 0.0241 | 29.0 | 2088 | 5.1367 | 0.3466 |
| 0.0197 | 30.0 | 2160 | 5.1914 | 0.3454 |
Framework versions
- Transformers 4.29.2
- Pytorch 2.0.1
- Datasets 2.12.0
- Tokenizers 0.13.3
- Task
- Text Generation
- Library
- transformers
- Type
- gpt_neo
- License
- mit
- Created
- Jul 19, 2023
- Updated
- Jul 19, 2023