metadata
library_name: transformers
license: mit
base_model: timpal0l/mdeberta-v3-base-squad2
tags:
- generated_from_trainer
model-index:
- name: my_awesome_qa_model_vifactcheck
results: []
my_awesome_qa_model_vifactcheck
This model is a fine-tuned version of timpal0l/mdeberta-v3-base-squad2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0136
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: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.3886 | 1.0 | 1037 | 0.2482 |
| 0.2857 | 2.0 | 2074 | 0.2373 |
| 0.2217 | 3.0 | 3111 | 0.1258 |
| 0.1717 | 4.0 | 4148 | 0.0892 |
| 0.1426 | 5.0 | 5185 | 0.0606 |
| 0.0924 | 6.0 | 6222 | 0.0524 |
| 0.071 | 7.0 | 7259 | 0.0270 |
| 0.0514 | 8.0 | 8296 | 0.0198 |
| 0.0348 | 9.0 | 9333 | 0.0148 |
| 0.024 | 10.0 | 10370 | 0.0136 |
Framework versions
- Transformers 4.52.4
- Pytorch 2.6.0+cu124
- Datasets 3.6.0
- Tokenizers 0.21.2