modernbert-classif-2

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: 1.0529
• F1: 0.8874

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: 5e-05
• train_batch_size: 128
• eval_batch_size: 16
• 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
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1
0.4048	1.0	359	0.3766	0.8496
0.2606	2.0	718	0.3401	0.8746
0.1339	3.0	1077	0.3965	0.8753
0.0794	4.0	1436	0.5419	0.8777
0.0621	5.0	1795	0.7029	0.8781
0.0447	6.0	2154	0.6386	0.8802
0.0363	7.0	2513	0.7959	0.8780
0.0321	8.0	2872	0.8826	0.8782
0.0209	9.0	3231	0.8348	0.8838
0.0174	10.0	3590	0.8174	0.8855
0.0148	11.0	3949	0.9351	0.8852
0.0106	12.0	4308	1.0183	0.8852
0.0109	13.0	4667	1.0120	0.8835
0.0095	14.0	5026	1.0379	0.8863
0.012	15.0	5385	1.0287	0.8871
0.0097	16.0	5744	1.0354	0.8862
0.0062	17.0	6103	1.0416	0.8881
0.0075	18.0	6462	1.0507	0.8874
0.0085	19.0	6821	1.0490	0.8879
0.0083	20.0	7180	1.0529	0.8874

Framework versions

• Transformers 4.48.0.dev0
• Pytorch 2.4.1+cu121
• Datasets 3.1.0
• Tokenizers 0.21.0