answerdotai-ModernBERT-base-anion-neg-bal-1e-06-256

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: 0.3196

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-06
• train_batch_size: 256
• eval_batch_size: 256
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss
0.6865	1.0	358	0.5785
0.5191	2.0	716	0.4656
0.435	3.0	1074	0.4231
0.3931	4.0	1432	0.3961
0.3758	5.0	1790	0.3774
0.3458	6.0	2148	0.3627
0.3247	7.0	2506	0.3526
0.3032	8.0	2864	0.3431
0.2864	9.0	3222	0.3359
0.2804	10.0	3580	0.3331
0.2644	11.0	3938	0.3240
0.254	12.0	4296	0.3206
0.2406	13.0	4654	0.3191
0.2324	14.0	5012	0.3188
0.2242	15.0	5370	0.3189
0.2139	16.0	5728	0.3159
0.2087	17.0	6086	0.3166
0.2019	18.0	6444	0.3151
0.1984	19.0	6802	0.3173
0.1887	20.0	7160	0.3203
0.1814	21.0	7518	0.3196

Framework versions

• Transformers 4.49.0
• Pytorch 2.6.0+cu124
• Datasets 3.3.2
• Tokenizers 0.21.1