library_name, license, base_model, tags, metrics, model-index
| library_name | license | base_model | tags | metrics | model-index | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| peft | apache-2.0 | bert-base-multilingual-cased |
|
|
|
mbert-loraxs-qa-vanilla
This model is a fine-tuned version of bert-base-multilingual-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.4437
- Exact Match: 55.68
- F1: 70.3954
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 0.1
- num_epochs: 7
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Exact Match | F1 |
|---|---|---|---|---|---|
| 1.6012 | 1.0 | 634 | 1.6034 | 52.8 | 66.3884 |
| 1.5347 | 2.0 | 1268 | 1.5703 | 54.16 | 67.8105 |
| 1.5039 | 3.0 | 1902 | 1.4684 | 55.12 | 69.1541 |
| 1.4706 | 4.0 | 2536 | 1.4466 | 55.68 | 69.7596 |
| 1.4197 | 5.0 | 3170 | 1.4419 | 56.48 | 70.7228 |
| 1.3971 | 6.0 | 3804 | 1.4516 | 55.76 | 70.6580 |
| 1.3627 | 7.0 | 4438 | 1.4437 | 55.68 | 70.3954 |
Framework versions
- PEFT 0.19.1
- Transformers 5.9.0
- Pytorch 2.11.0+cu128
- Datasets 4.8.5
- Tokenizers 0.22.2