eurobert-ner-model-20250726-134739

This model is a fine-tuned version of EuroBERT/EuroBERT-210m on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0803
• Precision: 0.8140
• Recall: 0.8593
• F1: 0.8360
• Accuracy: 0.9753

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: 0.0002
• train_batch_size: 128
• eval_batch_size: 128
• 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
• num_epochs: 7
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	0.4979	117	0.3960	0.2073	0.3370	0.2567	0.8704
No log	0.9957	234	0.1945	0.4331	0.6111	0.5069	0.9321
No log	1.4936	351	0.1142	0.6375	0.7556	0.6915	0.9554
No log	1.9915	468	0.0943	0.6603	0.7630	0.7079	0.9644
0.4252	2.4894	585	0.0852	0.7092	0.8037	0.7535	0.9677
0.4252	2.9872	702	0.0897	0.6926	0.7926	0.7392	0.9669
0.4252	3.4851	819	0.0949	0.7838	0.8593	0.8198	0.9747
0.4252	3.9830	936	0.0803	0.8140	0.8593	0.8360	0.9753
0.0446	4.4809	1053	0.1163	0.7441	0.8185	0.7795	0.9705
0.0446	4.9787	1170	0.1178	0.7690	0.8259	0.7964	0.9714
0.0446	5.4766	1287	0.1559	0.7736	0.8481	0.8092	0.9731
0.0446	5.9745	1404	0.1551	0.7416	0.8185	0.7782	0.9694
0.0123	6.4723	1521	0.2088	0.7363	0.7963	0.7651	0.9646
0.0123	6.9702	1638	0.2102	0.7491	0.8074	0.7772	0.9675

Framework versions

• Transformers 4.54.0
• Pytorch 2.7.1+cu126
• Datasets 4.0.0
• Tokenizers 0.21.2