results_bert

This model is a fine-tuned version of answerdotai/ModernBERT-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4209
• Accuracy: 0.8661

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: 512
• seed: 42
• optimizer: Use adamw_torch 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: 500
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	120	0.6514	0.6248
No log	2.0	240	0.5753	0.7198
No log	3.0	360	0.6240	0.6720
No log	4.0	480	0.4622	0.7847
0.6066	5.0	600	0.5026	0.7693
0.6066	6.0	720	0.4972	0.7817
0.6066	7.0	840	0.4209	0.7947
0.6066	8.0	960	0.4197	0.8142
0.4452	9.0	1080	0.5085	0.7935
0.4452	10.0	1200	0.3948	0.8265
0.4452	11.0	1320	0.3898	0.8307
0.4452	12.0	1440	0.3542	0.8490
0.322	13.0	1560	0.4070	0.8342
0.322	14.0	1680	0.3532	0.8454
0.322	15.0	1800	0.4472	0.8319
0.322	16.0	1920	0.3935	0.8549
0.2358	17.0	2040	0.3641	0.8625
0.2358	18.0	2160	0.3950	0.8614
0.2358	19.0	2280	0.4140	0.8655
0.2358	20.0	2400	0.4209	0.8661

Framework versions

• Transformers 4.53.0
• Pytorch 2.7.1+cu126
• Datasets 3.6.0
• Tokenizers 0.21.2