e93bcf3247289b9ad5d970633a50453d

This model is a fine-tuned version of FacebookAI/xlm-roberta-large-finetuned-conll02-spanish on the dim/tldr_news dataset. It achieves the following results on the evaluation set:

• Loss: 0.8507
• Data Size: 1.0
• Epoch Runtime: 33.2750
• Accuracy: 0.7649
• F1 Macro: 0.8035
• Rouge1: 0.7656
• Rouge2: 0.0
• Rougel: 0.7653
• Rougelsum: 0.7649

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: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• total_train_batch_size: 32
• total_eval_batch_size: 32
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: constant
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Data Size	Epoch Runtime	Accuracy	F1 Macro	Rouge1	Rouge2	Rougel	Rougelsum
No log	0	0	1.6917	0	2.6219	0.2486	0.0808	0.2486	0.0	0.2486	0.2482
No log	1	178	1.5785	0.0078	4.1250	0.2763	0.0876	0.2763	0.0	0.2756	0.2763
No log	2	356	1.3074	0.0156	3.7850	0.6080	0.4121	0.6087	0.0	0.6087	0.6072
No log	3	534	0.8929	0.0312	4.9785	0.5945	0.4358	0.5945	0.0	0.5945	0.5952
No log	4	712	0.8155	0.0625	6.6916	0.7188	0.5622	0.7202	0.0	0.7195	0.7195
No log	5	890	0.8059	0.125	8.7791	0.7259	0.5557	0.7266	0.0	0.7259	0.7251
0.0577	6	1068	1.0613	0.25	12.4444	0.6371	0.4495	0.6378	0.0	0.6371	0.6378
0.7498	7	1246	0.7125	0.5	19.1268	0.7457	0.5951	0.7468	0.0	0.7457	0.7457
0.6177	8.0	1424	0.6985	1.0	34.6716	0.7450	0.7396	0.7457	0.0	0.7457	0.7450
0.56	9.0	1602	0.6266	1.0	32.5332	0.7692	0.7949	0.7699	0.0	0.7699	0.7699
0.4602	10.0	1780	0.6521	1.0	32.9004	0.7585	0.7834	0.7592	0.0	0.7585	0.7592
0.408	11.0	1958	0.7225	1.0	32.9142	0.7422	0.7822	0.7422	0.0	0.7422	0.7422
0.3096	12.0	2136	0.7663	1.0	32.6184	0.7585	0.7985	0.7585	0.0	0.7592	0.7589
0.2925	13.0	2314	0.8507	1.0	33.2750	0.7649	0.8035	0.7656	0.0	0.7653	0.7649

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu128
• Datasets 4.3.0
• Tokenizers 0.22.1