metadata
license: cc-by-nc-4.0
tags:
- generated_from_trainer
datasets:
- squad_modified_for_t5_qg_2
model-index:
- name: greek-nllb-4ep-512
results: []
greek-nllb-4ep-512
This model is a fine-tuned version of facebook/nllb-200-distilled-600M on the squad_modified_for_t5_qg_2 dataset. It achieves the following results on the evaluation set:
- Loss: 1.2852
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 |
|---|---|---|---|
| 2.2208 | 0.17 | 100 | 1.5566 |
| 1.7198 | 0.34 | 200 | 1.4767 |
| 1.5435 | 0.51 | 300 | 1.4467 |
| 1.5017 | 0.67 | 400 | 1.3955 |
| 1.4569 | 0.84 | 500 | 1.3711 |
| 1.4496 | 1.01 | 600 | 1.3556 |
| 1.3174 | 1.18 | 700 | 1.3488 |
| 1.3025 | 1.35 | 800 | 1.3398 |
| 1.3013 | 1.52 | 900 | 1.3286 |
| 1.2938 | 1.69 | 1000 | 1.3161 |
| 1.2984 | 1.86 | 1100 | 1.3057 |
| 1.2494 | 2.03 | 1200 | 1.3097 |
| 1.1843 | 2.2 | 1300 | 1.2985 |
| 1.1868 | 2.36 | 1400 | 1.3021 |
| 1.1875 | 2.53 | 1500 | 1.2958 |
| 1.1854 | 2.7 | 1600 | 1.2987 |
| 1.1818 | 2.87 | 1700 | 1.2885 |
| 1.1696 | 3.04 | 1800 | 1.2912 |
| 1.1108 | 3.21 | 1900 | 1.2880 |
| 1.1301 | 3.38 | 2000 | 1.2889 |
| 1.1139 | 3.55 | 2100 | 1.2869 |
| 1.1098 | 3.72 | 2200 | 1.2879 |
| 1.1073 | 3.88 | 2300 | 1.2852 |
Framework versions
- Transformers 4.27.0.dev0
- Pytorch 1.13.1+cu116
- Datasets 2.10.1
- Tokenizers 0.13.2