dc22db767834880b27c3b7580d7f0e11

This model is a fine-tuned version of albert/albert-large-v2 on the dim/tldr_news dataset. It achieves the following results on the evaluation set:

• Loss: 0.7312
• Data Size: 1.0
• Epoch Runtime: 13.5041
• Accuracy: 0.7571
• F1 Macro: 0.7977
• Rouge1: 0.7578
• Rouge2: 0.0
• Rougel: 0.7578
• Rougelsum: 0.7571

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	1.6847	0	1.5042	0.2273	0.0883	0.2280	0.0	0.2280	0.2273
No log	1	178	1.4744	0.0078	2.5043	0.3430	0.1799	0.3438	0.0	0.3438	0.3430
No log	2	356	1.3236	0.0156	1.7831	0.4979	0.3396	0.4986	0.0	0.4986	0.4979
No log	3	534	1.1315	0.0312	1.8992	0.5114	0.3249	0.5114	0.0	0.5121	0.5114
No log	4	712	1.0243	0.0625	2.3969	0.6378	0.4864	0.6392	0.0	0.6385	0.6378
No log	5	890	0.8653	0.125	3.1190	0.6570	0.4593	0.6577	0.0	0.6577	0.6570
0.0671	6	1068	0.9167	0.25	4.5123	0.6776	0.5398	0.6783	0.0	0.6783	0.6776
0.7706	7	1246	0.7597	0.5	7.5424	0.7159	0.5701	0.7166	0.0	0.7159	0.7166
0.6373	8.0	1424	0.6951	1.0	13.6232	0.7259	0.7371	0.7266	0.0	0.7266	0.7266
0.603	9.0	1602	0.5889	1.0	13.3509	0.7699	0.8055	0.7706	0.0	0.7706	0.7706
0.5294	10.0	1780	0.6513	1.0	13.4181	0.7578	0.7974	0.7578	0.0	0.7578	0.7585
0.4812	11.0	1958	0.6771	1.0	13.6171	0.7479	0.7803	0.7479	0.0	0.7472	0.7479
0.3728	12.0	2136	0.6993	1.0	13.4518	0.7493	0.7976	0.7493	0.0	0.7493	0.75
0.3305	13.0	2314	0.7312	1.0	13.5041	0.7571	0.7977	0.7578	0.0	0.7578	0.7571

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.1