modernbert_clincl_oos

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: 0.1483
• Accuracy: 0.9703
• F1: 0.9699

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: 1e-05
• train_batch_size: 32
• eval_batch_size: 32
• 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: cosine
• lr_scheduler_warmup_ratio: 0.06
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
4.1273	0.4193	200	1.4172	0.6848	0.6712
0.4889	0.8386	400	0.2904	0.9306	0.9299
0.1411	1.2579	600	0.2075	0.9503	0.9502
0.083	1.6771	800	0.1853	0.9577	0.9573
0.0659	2.0964	1000	0.1509	0.9706	0.9704
0.0235	2.5157	1200	0.1619	0.9626	0.9623
0.0117	2.9350	1400	0.1464	0.9697	0.9693
0.0066	3.3543	1600	0.1516	0.9684	0.9679
0.0011	3.7736	1800	0.1478	0.9703	0.9699
0.0051	4.1929	2000	0.1477	0.9697	0.9693
0.0013	4.6122	2200	0.1483	0.9703	0.9699

Framework versions

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