greek-nllb-4ep-384
This model is a fine-tuned version of facebook/nllb-200-distilled-600M on the squad dataset. It achieves the following results on the evaluation set:
- Loss: 1.2875
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: 0.0001
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.215 | 0.17 | 100 | 1.5536 |
| 1.7149 | 0.34 | 200 | 1.4819 |
| 1.544 | 0.51 | 300 | 1.4381 |
| 1.5006 | 0.67 | 400 | 1.3999 |
| 1.4577 | 0.84 | 500 | 1.3756 |
| 1.4495 | 1.01 | 600 | 1.3613 |
| 1.32 | 1.18 | 700 | 1.3467 |
| 1.2999 | 1.35 | 800 | 1.3404 |
| 1.2993 | 1.52 | 900 | 1.3339 |
| 1.2909 | 1.69 | 1000 | 1.3189 |
| 1.2974 | 1.86 | 1100 | 1.3112 |
| 1.2516 | 2.03 | 1200 | 1.3171 |
| 1.1852 | 2.2 | 1300 | 1.3032 |
| 1.1862 | 2.36 | 1400 | 1.3072 |
| 1.184 | 2.53 | 1500 | 1.2967 |
| 1.1865 | 2.7 | 1600 | 1.2968 |
| 1.1797 | 2.87 | 1700 | 1.2903 |
| 1.1707 | 3.04 | 1800 | 1.2934 |
| 1.1128 | 3.21 | 1900 | 1.2927 |
| 1.1314 | 3.38 | 2000 | 1.2895 |
| 1.1141 | 3.55 | 2100 | 1.2889 |
| 1.1137 | 3.72 | 2200 | 1.2888 |
| 1.1069 | 3.88 | 2300 | 1.2875 |
Framework versions
- Transformers 4.27.0.dev0
- Pytorch 1.13.0
- Datasets 2.1.0
- Tokenizers 0.13.2