eurobert210m_RSE_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: 0.0069
• Accuracy: 0.9982
• F1: 0.9982

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
0.7448	1.0	138	0.2380	0.9194	0.9200
0.3157	2.0	276	0.1846	0.9421	0.9419
0.2241	3.0	414	0.1905	0.9373	0.9371
0.1923	4.0	552	0.0821	0.9739	0.9739
0.1312	5.0	690	0.1449	0.9614	0.9616
0.1418	6.0	828	0.0782	0.9796	0.9795
0.1008	7.0	966	0.0579	0.9877	0.9877
0.0981	8.0	1104	0.0363	0.9893	0.9893
0.0723	9.0	1242	0.1002	0.9789	0.9789
0.0846	10.0	1380	0.0457	0.9907	0.9907
0.0779	11.0	1518	0.0620	0.9880	0.9880
0.0676	12.0	1656	0.0314	0.9932	0.9932
0.0389	13.0	1794	0.0232	0.9950	0.9950
0.0453	14.0	1932	0.0145	0.9966	0.9966
0.0328	15.0	2070	0.0303	0.9936	0.9936
0.0316	16.0	2208	0.0247	0.9948	0.9948
0.0191	17.0	2346	0.0070	0.9984	0.9984
0.0209	18.0	2484	0.0069	0.9982	0.9982

Framework versions

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