arbitration_best_model_modernbert

This model is a fine-tuned version of answerdotai/ModernBERT-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2131
• Accuracy: 0.9607
• Precision: 0.9067
• Recall: 1.0
• F1: 0.9510

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: 8e-05
• train_batch_size: 32
• eval_batch_size: 64
• 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: 100
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.5125	1.0	50	0.2728	0.9101	0.8824	0.8824	0.8824
0.4129	2.0	100	0.3523	0.8989	0.7976	0.9853	0.8816
0.2913	3.0	150	0.5966	0.8876	0.9615	0.7353	0.8333
0.1508	4.0	200	0.1999	0.9326	0.9118	0.9118	0.9118
0.1175	5.0	250	0.1692	0.9438	0.9394	0.9118	0.9254
0.0977	6.0	300	0.1286	0.9551	0.8947	1.0	0.9444
0.0531	7.0	350	0.1609	0.9494	0.9275	0.9412	0.9343
0.1658	8.0	400	0.1969	0.9494	0.9041	0.9706	0.9362
0.0388	9.0	450	0.2131	0.9607	0.9067	1.0	0.9510
0.0944	10.0	500	0.2231	0.9494	0.9155	0.9559	0.9353
0.0659	11.0	550	0.2025	0.9607	0.9067	1.0	0.9510
0.0201	12.0	600	0.1815	0.9551	0.9054	0.9853	0.9437
0.0404	13.0	650	0.2307	0.9494	0.9041	0.9706	0.9362
0.0157	14.0	700	0.2647	0.9551	0.9054	0.9853	0.9437

Framework versions

• Transformers 5.15.1
• Pytorch 2.11.0+cu128
• Datasets 5.0.1
• Tokenizers 0.22.2