mb-bert_results

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

• Loss: 0.4420
• Accuracy: 0.8726
• Precision: 0.8768
• Recall: 0.8669
• F1-score: 0.8718

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: 2e-05
• train_batch_size: 4
• eval_batch_size: 4
• 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: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1-score
0.2678	1.0	6250	0.5000	0.8542	0.8991	0.7978	0.8455
0.5841	2.0	12500	0.4420	0.8726	0.8768	0.8669	0.8718
0.1813	3.0	18750	0.4602	0.8745	0.8611	0.8930	0.8768

Framework versions

• PEFT 0.14.0
• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.3.2
• Tokenizers 0.21.0