USS-reward-model-hybrid
This model is a fine-tuned version of answerdotai/ModernBERT-large on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.2303
- Mse: 1.1759
- Mae: 0.9844
- R2: -5.0964
- Spearman Correlation: -0.0139
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: 2
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- gradient_accumulation_steps: 10
- total_train_batch_size: 20
- 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
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Mse | Mae | R2 | Spearman Correlation |
|---|---|---|---|---|---|---|---|
| 12.2907 | 1.0 | 97 | 1.2303 | 1.1759 | 0.9844 | -5.0964 | -0.0139 |
| 12.2915 | 2.0 | 194 | 1.2303 | 1.1759 | 0.9844 | -5.0964 | -0.0139 |
| 12.2892 | 3.0 | 291 | 1.2303 | 1.1759 | 0.9844 | -5.0964 | -0.0139 |
| 12.2928 | 4.0 | 388 | 1.2303 | 1.1759 | 0.9844 | -5.0964 | -0.0139 |
| 12.2896 | 5.0 | 485 | 1.2303 | 1.1759 | 0.9844 | -5.0964 | -0.0139 |
| 12.2913 | 6.0 | 582 | 1.2303 | 1.1759 | 0.9844 | -5.0964 | -0.0139 |
| 12.2916 | 7.0 | 679 | 1.2303 | 1.1759 | 0.9844 | -5.0964 | -0.0139 |
| 12.2936 | 8.0 | 776 | 1.2303 | 1.1759 | 0.9844 | -5.0964 | -0.0139 |
| 12.2891 | 9.0 | 873 | 1.2303 | 1.1759 | 0.9844 | -5.0964 | -0.0139 |
| 12.2917 | 10.0 | 970 | 1.2303 | 1.1759 | 0.9844 | -5.0964 | -0.0139 |
Framework versions
- Transformers 5.9.0
- Pytorch 2.12.0+cu130
- Datasets 4.8.5
- Tokenizers 0.22.2
- Downloads last month
- 5
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for athirorg/USS-reward-model-hybrid
Base model
answerdotai/ModernBERT-large