|
--- |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: vicuna-adv-robust-u50-sft-lora |
|
results: [] |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# vicuna-adv-robust-u50-sft-lora |
|
|
|
This model was trained from scratch on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.2125 |
|
|
|
## 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: 0.0003 |
|
- train_batch_size: 8 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- distributed_type: multi-GPU |
|
- num_devices: 4 |
|
- gradient_accumulation_steps: 16 |
|
- total_train_batch_size: 512 |
|
- total_eval_batch_size: 32 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- num_epochs: 50 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| No log | 0 | 0 | 2.4952 | |
|
| 2.5615 | 1.09 | 1 | 2.5270 | |
|
| 2.5615 | 1.09 | 1 | 2.5362 | |
|
| 2.5615 | 3.03 | 2 | 2.5342 | |
|
| 2.5615 | 4.12 | 3 | 2.2735 | |
|
| 2.5615 | 4.12 | 3 | 2.3209 | |
|
| 2.5615 | 6.06 | 4 | 2.1017 | |
|
| 2.363 | 7.15 | 5 | 2.0121 | |
|
| 2.363 | 7.15 | 5 | 2.0751 | |
|
| 2.363 | 9.09 | 6 | 1.9646 | |
|
| 2.363 | 9.09 | 6 | 1.8912 | |
|
| 2.363 | 11.03 | 7 | 1.8100 | |
|
| 2.363 | 12.12 | 8 | 1.8144 | |
|
| 2.363 | 12.12 | 8 | 1.7983 | |
|
| 2.363 | 14.06 | 9 | 1.7634 | |
|
| 1.9009 | 15.15 | 10 | 1.7628 | |
|
| 1.9009 | 15.15 | 10 | 1.7354 | |
|
| 1.9009 | 17.09 | 11 | 1.7343 | |
|
| 1.9009 | 17.09 | 11 | 1.7232 | |
|
| 1.9009 | 19.03 | 12 | 1.6737 | |
|
| 1.9009 | 20.12 | 13 | 1.6418 | |
|
| 1.9009 | 20.12 | 13 | 1.6635 | |
|
| 1.9009 | 22.06 | 14 | 1.6280 | |
|
| 1.7031 | 23.15 | 15 | 1.6042 | |
|
| 1.7031 | 23.15 | 15 | 1.6120 | |
|
| 1.7031 | 25.09 | 16 | 1.5792 | |
|
| 1.7031 | 25.09 | 16 | 1.6128 | |
|
| 1.7031 | 27.03 | 17 | 1.5468 | |
|
| 1.7031 | 28.12 | 18 | 1.5303 | |
|
| 1.7031 | 28.12 | 18 | 1.5160 | |
|
| 1.7031 | 30.06 | 19 | 1.5195 | |
|
| 1.5968 | 31.15 | 20 | 1.5098 | |
|
| 1.5968 | 31.15 | 20 | 1.4775 | |
|
| 1.5968 | 33.09 | 21 | 1.4770 | |
|
| 1.5968 | 33.09 | 21 | 1.4588 | |
|
| 1.5968 | 35.03 | 22 | 1.4474 | |
|
| 1.5968 | 36.12 | 23 | 1.4240 | |
|
| 1.5968 | 36.12 | 23 | 1.4164 | |
|
| 1.5968 | 38.06 | 24 | 1.4060 | |
|
| 1.4776 | 39.15 | 25 | 1.3753 | |
|
| 1.4776 | 39.15 | 25 | 1.3858 | |
|
| 1.4776 | 41.09 | 26 | 1.3822 | |
|
| 1.4776 | 41.09 | 26 | 1.3268 | |
|
| 1.4776 | 43.03 | 27 | 1.3443 | |
|
| 1.4776 | 44.12 | 28 | 1.3259 | |
|
| 1.4776 | 44.12 | 28 | 1.3117 | |
|
| 1.4776 | 46.06 | 29 | 1.3105 | |
|
| 1.3585 | 47.15 | 30 | 1.2553 | |
|
| 1.3585 | 47.15 | 30 | 1.2755 | |
|
| 1.3585 | 49.09 | 31 | 1.2036 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.35.0 |
|
- Pytorch 2.1.0a0+32f93b1 |
|
- Datasets 2.14.6 |
|
- Tokenizers 0.14.1 |
|
|
