opt-350m_lora_lr5e-05_bs4_epoch20_wd0.01
This model is a fine-tuned version of facebook/opt-350m on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.4811
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: 5e-05
- train_batch_size: 4
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 3.9852 | 1.0 | 157 | 3.5147 |
| 3.5297 | 2.0 | 314 | 3.4018 |
| 3.4497 | 3.0 | 471 | 3.3918 |
| 3.3979 | 4.0 | 628 | 3.4111 |
| 3.3625 | 5.0 | 785 | 3.3598 |
| 3.3397 | 6.0 | 942 | 3.3665 |
| 3.2897 | 7.0 | 1099 | 3.3855 |
| 3.2289 | 8.0 | 1256 | 3.3887 |
| 3.287 | 9.0 | 1413 | 3.3999 |
| 3.1938 | 10.0 | 1570 | 3.4151 |
| 3.1485 | 11.0 | 1727 | 3.4140 |
| 3.0905 | 12.0 | 1884 | 3.4125 |
| 3.0712 | 13.0 | 2041 | 3.4385 |
| 3.0604 | 14.0 | 2198 | 3.4475 |
| 3.0445 | 15.0 | 2355 | 3.4718 |
| 2.9898 | 16.0 | 2512 | 3.4659 |
| 2.9948 | 17.0 | 2669 | 3.4765 |
| 2.984 | 18.0 | 2826 | 3.4740 |
| 3.0367 | 19.0 | 2983 | 3.4817 |
| 2.9656 | 20.0 | 3140 | 3.4811 |
Framework versions
- PEFT 0.7.1
- Transformers 4.36.2
- Pytorch 2.0.1
- Datasets 2.16.1
- Tokenizers 0.15.0
- Downloads last month
- 2
Model tree for jysssacc/opt-350m_lora_lr5e-05_bs4_epoch20_wd0.01
Base model
facebook/opt-350m