results

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: 0.2000
• Accuracy: 0.9433
• F1: 0.9429
• Precision: 0.9508
• Recall: 0.9433

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: 4e-05
• train_batch_size: 6
• eval_batch_size: 6
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 12
• 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: 2
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
3.8028	0.0833	25	1.5191	0.3944	0.2893	0.4598	0.3944
2.2046	0.1667	50	0.7147	0.75	0.7423	0.7685	0.75
1.2172	0.25	75	0.6074	0.7989	0.7727	0.8508	0.7989
0.9054	0.3333	100	0.3817	0.8656	0.8637	0.8907	0.8656
0.873	0.4167	125	0.3460	0.8678	0.8665	0.8810	0.8678
0.7074	0.5	150	0.2918	0.8889	0.8848	0.9159	0.8889
1.0552	0.5833	175	0.2550	0.89	0.8868	0.9130	0.89
0.5167	0.6667	200	0.2660	0.9044	0.9043	0.9071	0.9044
0.3174	0.75	225	0.2641	0.8956	0.8882	0.9235	0.8956
0.3369	0.8333	250	0.1745	0.9489	0.9490	0.9520	0.9489
0.2966	0.9167	275	0.1484	0.9567	0.9568	0.9589	0.9567
0.5544	1.0	300	0.2000	0.9433	0.9429	0.9508	0.9433

Framework versions

• Transformers 4.57.1
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1