metadata
tags:
- generated_from_trainer
model-index:
- name: deberta-base.as.finetuned
results: []
deberta-base.as.finetuned
This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 5.6126
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 30
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 6.6155 | 0.47 | 1000 | 6.5530 |
| 6.4654 | 0.93 | 2000 | 6.4358 |
| 6.3684 | 1.4 | 3000 | 6.3436 |
| 6.2843 | 1.87 | 4000 | 6.2675 |
| 6.2171 | 2.33 | 5000 | 6.2000 |
| 6.1687 | 2.8 | 6000 | 6.1522 |
| 6.122 | 3.27 | 7000 | 6.1094 |
| 6.0828 | 3.73 | 8000 | 6.0599 |
| 6.0346 | 4.2 | 9000 | 6.0132 |
| 5.9935 | 4.67 | 10000 | 5.9940 |
| 5.9553 | 5.13 | 11000 | 5.9589 |
| 5.9347 | 5.6 | 12000 | 5.9315 |
| 5.9105 | 6.07 | 13000 | 5.8996 |
| 5.8851 | 6.53 | 14000 | 5.8859 |
| 5.8632 | 7.0 | 15000 | 5.8626 |
| 5.847 | 7.47 | 16000 | 5.8495 |
| 5.8393 | 7.93 | 17000 | 5.8276 |
| 5.8041 | 8.4 | 18000 | 5.8096 |
| 5.7922 | 8.87 | 19000 | 5.7968 |
| 5.7758 | 9.33 | 20000 | 5.7869 |
| 5.7656 | 9.8 | 21000 | 5.7795 |
| 5.7533 | 10.27 | 22000 | 5.7652 |
| 5.7484 | 10.73 | 23000 | 5.7511 |
| 5.7352 | 11.2 | 24000 | 5.7417 |
| 5.7215 | 11.67 | 25000 | 5.7327 |
| 5.7126 | 12.13 | 26000 | 5.7352 |
| 5.7048 | 12.6 | 27000 | 5.7233 |
| 5.6887 | 13.07 | 28000 | 5.7106 |
| 5.6892 | 13.53 | 29000 | 5.7003 |
| 5.6819 | 14.0 | 30000 | 5.6993 |
| 5.6665 | 14.47 | 31000 | 5.7001 |
| 5.6704 | 14.93 | 32000 | 5.6936 |
| 5.6608 | 15.4 | 33000 | 5.6828 |
| 5.6575 | 15.87 | 34000 | 5.6795 |
| 5.6552 | 16.33 | 35000 | 5.6686 |
| 5.647 | 16.8 | 36000 | 5.6676 |
| 5.639 | 17.27 | 37000 | 5.6659 |
| 5.6346 | 17.73 | 38000 | 5.6644 |
| 5.633 | 18.2 | 39000 | 5.6648 |
| 5.6276 | 18.67 | 40000 | 5.6569 |
| 5.6252 | 19.13 | 41000 | 5.6488 |
| 5.6254 | 19.6 | 42000 | 5.6468 |
| 5.6198 | 20.07 | 43000 | 5.6485 |
| 5.6104 | 20.53 | 44000 | 5.6394 |
| 5.6117 | 21.0 | 45000 | 5.6417 |
| 5.6048 | 21.47 | 46000 | 5.6362 |
| 5.5973 | 21.93 | 47000 | 5.6378 |
| 5.5923 | 22.4 | 48000 | 5.6312 |
| 5.6047 | 22.87 | 49000 | 5.6308 |
| 5.5955 | 23.33 | 50000 | 5.6341 |
| 5.5947 | 23.8 | 51000 | 5.6257 |
| 5.5979 | 24.27 | 52000 | 5.6330 |
| 5.5924 | 24.73 | 53000 | 5.6200 |
| 5.5996 | 25.2 | 54000 | 5.6175 |
| 5.5823 | 25.66 | 55000 | 5.6245 |
| 5.5884 | 26.13 | 56000 | 5.6207 |
| 5.5828 | 26.6 | 57000 | 5.6184 |
| 5.5844 | 27.06 | 58000 | 5.6255 |
| 5.5793 | 27.53 | 59000 | 5.6269 |
| 5.5809 | 28.0 | 60000 | 5.6185 |
| 5.5786 | 28.46 | 61000 | 5.6185 |
| 5.5765 | 28.93 | 62000 | 5.6204 |
| 5.5814 | 29.4 | 63000 | 5.6257 |
| 5.5837 | 29.86 | 64000 | 5.6163 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu118
- Datasets 2.17.1
- Tokenizers 0.15.2