loha_fine_tuned_copa_croslo
This model is a fine-tuned version of EMBEDDIA/crosloengual-bert on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.2439
- Accuracy: 0.64
- F1: 0.6409
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: 0.003
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- training_steps: 400
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 0.693 | 1.0 | 50 | 0.6764 | 0.56 | 0.56 |
| 0.6401 | 2.0 | 100 | 0.6814 | 0.53 | 0.5304 |
| 0.5345 | 3.0 | 150 | 0.6725 | 0.66 | 0.6592 |
| 0.3293 | 4.0 | 200 | 0.8363 | 0.65 | 0.6509 |
| 0.1831 | 5.0 | 250 | 0.9311 | 0.65 | 0.6507 |
| 0.0662 | 6.0 | 300 | 1.0890 | 0.67 | 0.6707 |
| 0.0238 | 7.0 | 350 | 1.1631 | 0.64 | 0.6409 |
| 0.0364 | 8.0 | 400 | 1.2439 | 0.64 | 0.6409 |
Framework versions
- PEFT 0.11.1
- Transformers 4.40.2
- Pytorch 2.1.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1
- Downloads last month
- 3
Model tree for anzeo/loha_fine_tuned_copa_croslo
Base model
EMBEDDIA/crosloengual-bert