b0ef99bd633b80a07ed4b29b0a2c5d91

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

• Loss: 1.4613
• Data Size: 1.0
• Epoch Runtime: 25.1440
• Accuracy: 0.2756
• F1 Macro: 0.0864
• Rouge1: 0.2752
• Rouge2: 0.0
• Rougel: 0.2749
• Rougelsum: 0.2756

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	1.7195	0	2.3979	0.2173	0.0714	0.2166	0.0	0.2173	0.2173
No log	1	178	1.6290	0.0078	3.4422	0.2472	0.0793	0.2472	0.0	0.2472	0.2464
No log	2	356	1.6889	0.0156	2.8855	0.0987	0.0825	0.0987	0.0	0.0987	0.0987
No log	3	534	1.4780	0.0312	3.2746	0.2479	0.1712	0.2479	0.0	0.2472	0.2479
No log	4	712	1.5387	0.0625	4.0645	0.2756	0.0864	0.2752	0.0	0.2749	0.2756
No log	5	890	1.4859	0.125	5.4759	0.2756	0.0864	0.2752	0.0	0.2749	0.2756
0.0972	6	1068	1.4760	0.25	8.3402	0.2173	0.0714	0.2166	0.0	0.2173	0.2173
1.461	7	1246	1.4646	0.5	13.7672	0.2486	0.0796	0.2486	0.0	0.2486	0.2479
1.4427	8.0	1424	1.4623	1.0	25.4379	0.2756	0.0864	0.2752	0.0	0.2749	0.2756
1.4515	9.0	1602	1.4538	1.0	25.2654	0.2756	0.0864	0.2752	0.0	0.2749	0.2756
1.4354	10.0	1780	1.4681	1.0	25.2701	0.2486	0.0796	0.2486	0.0	0.2486	0.2479
1.4487	11.0	1958	1.4548	1.0	25.2500	0.2401	0.0774	0.2401	0.0	0.2408	0.2393
1.4558	12.0	2136	1.4588	1.0	25.2566	0.2401	0.0774	0.2401	0.0	0.2408	0.2393
1.4562	13.0	2314	1.4537	1.0	25.1478	0.2706	0.1402	0.2699	0.0	0.2692	0.2706
1.4467	14.0	2492	1.4537	1.0	25.1871	0.2173	0.0714	0.2166	0.0	0.2173	0.2173
1.4535	15.0	2670	1.4568	1.0	25.1471	0.2486	0.0796	0.2486	0.0	0.2486	0.2479
1.4578	16.0	2848	1.4544	1.0	25.1672	0.2756	0.0864	0.2752	0.0	0.2749	0.2756
1.4413	17.0	3026	1.4510	1.0	25.2195	0.2756	0.0864	0.2752	0.0	0.2749	0.2756
1.4454	18.0	3204	1.4521	1.0	25.1672	0.2486	0.0796	0.2486	0.0	0.2486	0.2479
1.4468	19.0	3382	1.4508	1.0	25.0699	0.2756	0.0864	0.2752	0.0	0.2749	0.2756
1.4568	20.0	3560	1.4548	1.0	25.2263	0.2756	0.0864	0.2752	0.0	0.2749	0.2756
1.4404	21.0	3738	1.4542	1.0	25.2249	0.2401	0.0774	0.2401	0.0	0.2408	0.2393
1.4448	22.0	3916	1.4589	1.0	25.3352	0.2486	0.0796	0.2486	0.0	0.2486	0.2479
1.4398	23.0	4094	1.4506	1.0	25.2263	0.2756	0.0864	0.2752	0.0	0.2749	0.2756
1.4484	24.0	4272	1.4517	1.0	25.2167	0.2486	0.0796	0.2486	0.0	0.2486	0.2479
1.4381	25.0	4450	1.4647	1.0	25.2744	0.2756	0.0864	0.2752	0.0	0.2749	0.2756
1.4482	26.0	4628	1.4524	1.0	25.5011	0.2756	0.0864	0.2752	0.0	0.2749	0.2756
1.4197	27.0	4806	1.4613	1.0	25.1440	0.2756	0.0864	0.2752	0.0	0.2749	0.2756

Framework versions

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