bert-conll2003-ner

This model is a fine-tuned version of bert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0514
• Precision: 0.9196
• Recall: 0.9414
• F1: 0.9304
• Accuracy: 0.9860

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_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 2
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.0703	0.2278	100	0.0731	0.8696	0.9150	0.8918	0.9794
0.0895	0.4556	200	0.0585	0.8960	0.9249	0.9102	0.9824
0.0699	0.6834	300	0.0571	0.9067	0.9268	0.9166	0.9835
0.0636	0.9112	400	0.0493	0.9212	0.9391	0.9301	0.9858
0.0405	1.1390	500	0.0525	0.9232	0.9428	0.9329	0.9862
0.0293	1.3667	600	0.0556	0.9251	0.9399	0.9325	0.9859
0.0307	1.5945	700	0.0516	0.9220	0.9408	0.9313	0.9858
0.025	1.8223	800	0.0514	0.9196	0.9414	0.9304	0.9860

Framework versions

• Transformers 4.55.4
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.21.4