eurobert210m_Eau_v2

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: 0.0680
• Accuracy: 0.9584
• F1: 0.9595

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
1.4372	1.0	67	0.9689	0.6322	0.5664
0.8205	2.0	134	0.6235	0.8213	0.8222
0.4899	3.0	201	0.4782	0.8326	0.8367
0.3598	4.0	268	0.2252	0.9196	0.9200
0.2854	5.0	335	0.2137	0.9258	0.9265
0.2054	6.0	402	0.1284	0.9452	0.9443
0.1735	7.0	469	0.1984	0.9296	0.9303
0.1763	8.0	536	0.1177	0.9409	0.9379
0.1601	9.0	603	0.1133	0.9485	0.9462
0.1206	10.0	670	0.1219	0.9461	0.9448
0.1269	11.0	737	0.0756	0.9565	0.9575
0.1238	12.0	804	0.1025	0.9522	0.9539
0.0969	13.0	871	0.0823	0.9570	0.9580
0.1046	14.0	938	0.0802	0.9527	0.9513
0.1101	15.0	1005	0.0797	0.9546	0.9539
0.0864	16.0	1072	0.0853	0.9565	0.9550
0.1002	17.0	1139	0.0696	0.9579	0.9582
0.0794	18.0	1206	0.0774	0.9579	0.9588
0.0849	19.0	1273	0.0719	0.9546	0.9529
0.0867	20.0	1340	0.0723	0.9589	0.9575
0.0952	21.0	1407	0.0680	0.9584	0.9595

Framework versions

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