JT_RV3_model

This model is a fine-tuned version of EMBEDDIA/crosloengual-bert on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2677
• Precision: 0.9469
• Recall: 0.9474
• F1: 0.9471
• Accuracy: 0.9569

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: 16
• eval_batch_size: 16
• 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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.2244	1.0	1905	0.1890	0.9381	0.9376	0.9379	0.9489
0.1454	2.0	3810	0.1732	0.9428	0.9435	0.9432	0.9534
0.1061	3.0	5715	0.1728	0.9460	0.9460	0.9460	0.9559
0.0783	4.0	7620	0.1864	0.9465	0.9462	0.9464	0.9563
0.0658	5.0	9525	0.1990	0.9449	0.9442	0.9445	0.9550
0.0409	6.0	11430	0.2166	0.9461	0.9465	0.9463	0.9564
0.0329	7.0	13335	0.2382	0.9459	0.9466	0.9462	0.9561
0.0251	8.0	15240	0.2519	0.9464	0.9467	0.9466	0.9564
0.0195	9.0	17145	0.2625	0.9462	0.9473	0.9467	0.9566
0.0144	10.0	19050	0.2677	0.9469	0.9474	0.9471	0.9569

Framework versions

• Transformers 5.0.0
• Pytorch 2.10.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2