432b938d256991d1d48dbe465de15232

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

• Loss: 1.0195
• Data Size: 1.0
• Epoch Runtime: 17.4149
• Accuracy: 0.7337
• F1 Macro: 0.7760
• Rouge1: 0.7344
• Rouge2: 0.0
• Rougel: 0.7344
• Rougelsum: 0.7337

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	1.6653	0	1.6907	0.2173	0.0714	0.2166	0.0	0.2173	0.2173
No log	1	178	1.6227	0.0078	2.7814	0.2173	0.0714	0.2166	0.0	0.2173	0.2173
No log	2	356	1.5512	0.0156	2.1478	0.2173	0.0716	0.2166	0.0	0.2173	0.2173
No log	3	534	1.2955	0.0312	2.8145	0.4354	0.2352	0.4354	0.0	0.4361	0.4347
No log	4	712	0.9631	0.0625	3.7560	0.6477	0.4707	0.6491	0.0	0.6484	0.6477
No log	5	890	0.8533	0.125	4.7372	0.6832	0.4976	0.6832	0.0	0.6839	0.6832
0.0641	6	1068	0.7924	0.25	6.7494	0.7102	0.5682	0.7109	0.0	0.7109	0.7102
0.6691	7	1246	0.7863	0.5	10.1592	0.7230	0.5544	0.7244	0.0	0.7237	0.7223
0.5934	8.0	1424	0.7012	1.0	18.0215	0.7358	0.5630	0.7365	0.0	0.7365	0.7358
0.4999	9.0	1602	0.6185	1.0	17.2786	0.7621	0.7567	0.7621	0.0	0.7628	0.7614
0.3875	10.0	1780	0.7163	1.0	17.1144	0.7450	0.7354	0.7454	0.0	0.7457	0.7443
0.3002	11.0	1958	0.7808	1.0	18.2779	0.7429	0.7576	0.7436	0.0	0.7436	0.7429
0.2323	12.0	2136	0.8669	1.0	16.8135	0.7621	0.7718	0.7621	0.0	0.7628	0.7621
0.1398	13.0	2314	1.0195	1.0	17.4149	0.7337	0.7760	0.7344	0.0	0.7344	0.7337

Framework versions

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