b770b41b49126880ea34b963e1b495ae

This model is a fine-tuned version of distilbert/distilbert-base-uncased-distilled-squad on the nyu-mll/glue [mrpc] dataset. It achieves the following results on the evaluation set:

• Loss: 1.0880
• Data Size: 1.0
• Epoch Runtime: 4.5971
• Accuracy: 0.7995
• F1 Macro: 0.7422
• Rouge1: 0.8001
• Rouge2: 0.0
• Rougel: 0.7995
• Rougelsum: 0.8001

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	0.7205	0	1.0356	0.3349	0.2509	0.3343	0.0	0.3355	0.3349
No log	1	114	0.6456	0.0078	1.7553	0.6651	0.3994	0.6657	0.0	0.6645	0.6651
No log	2	228	0.6898	0.0156	1.3485	0.6651	0.3994	0.6657	0.0	0.6645	0.6651
No log	3	342	0.6489	0.0312	1.3378	0.6651	0.3994	0.6657	0.0	0.6645	0.6651
0.0205	4	456	0.5963	0.0625	1.6037	0.7052	0.5653	0.7046	0.0	0.7046	0.7052
0.0205	5	570	0.5801	0.125	1.8321	0.7070	0.5362	0.7070	0.0	0.7064	0.7070
0.0205	6	684	0.5092	0.25	2.4145	0.7577	0.7133	0.7571	0.0	0.7583	0.7577
0.1301	7	798	0.4574	0.5	2.8118	0.7983	0.7537	0.7989	0.0	0.7983	0.7989
0.3348	8.0	912	0.4280	1.0	4.6597	0.8231	0.7883	0.8231	0.0	0.8231	0.8231
0.1845	9.0	1026	0.6303	1.0	4.3844	0.8149	0.7843	0.8154	0.0	0.8149	0.8149
0.0995	10.0	1140	0.7343	1.0	4.3817	0.8019	0.7727	0.8019	0.0	0.8025	0.8019
0.081	11.0	1254	0.8451	1.0	4.6516	0.8154	0.7986	0.8154	0.0	0.8154	0.8154
0.0569	12.0	1368	1.0880	1.0	4.5971	0.7995	0.7422	0.8001	0.0	0.7995	0.8001

Framework versions

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