metadata
license: mit
base_model: microsoft/deberta-v3-base
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: deberta-v3-base-finetuned-m_avoid_harm_seler
results: []
deberta-v3-base-finetuned-m_avoid_harm_seler
This model is a fine-tuned version of microsoft/deberta-v3-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5284
- Accuracy: 0.88
- F1: 0.9094
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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| 0.6798 | 1.0 | 26 | 0.6119 | 0.945 | 0.9377 |
| 0.5411 | 2.0 | 52 | 0.4665 | 0.78 | 0.8484 |
| 0.4348 | 3.0 | 78 | 0.4732 | 0.71 | 0.8012 |
| 0.2908 | 4.0 | 104 | 0.3952 | 0.855 | 0.8943 |
| 0.2341 | 5.0 | 130 | 0.5284 | 0.88 | 0.9094 |
Framework versions
- Transformers 4.41.1
- Pytorch 2.3.0+cu121
- Tokenizers 0.19.1