eurobert210m_Energie_v1

This model is a fine-tuned version of EuroBERT/EuroBERT-210m on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0418
• Accuracy: 0.9900
• F1: 0.9900

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: 32
• eval_batch_size: 32
• 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: 100
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
1.4438	1.0	116	0.9257	0.6453	0.5924
0.7711	2.0	232	0.5221	0.8263	0.8141
0.3803	3.0	348	0.1986	0.9357	0.9330
0.2678	4.0	464	0.1487	0.9501	0.9508
0.1839	5.0	580	0.1222	0.9642	0.9642
0.1736	6.0	696	0.1348	0.9634	0.9638
0.1207	7.0	812	0.0549	0.9829	0.9828
0.1022	8.0	928	0.0642	0.9802	0.9804
0.0699	9.0	1044	0.0723	0.9777	0.9773
0.0584	10.0	1160	0.0315	0.9886	0.9886
0.0617	11.0	1276	0.0506	0.9862	0.9861
0.0591	12.0	1392	0.0445	0.9867	0.9867
0.0599	13.0	1508	0.0402	0.9870	0.9870
0.0595	14.0	1624	0.0952	0.9788	0.9791
0.0401	15.0	1740	0.0526	0.9853	0.9854
0.0424	16.0	1856	0.0305	0.9902	0.9902
0.0323	17.0	1972	0.0969	0.9802	0.9798
0.0664	18.0	2088	0.0359	0.9881	0.9879
0.031	19.0	2204	0.0372	0.9897	0.9897
0.0239	20.0	2320	0.0418	0.9900	0.9900

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.3.2
• Tokenizers 0.21.0