43d4aa381d5eedf1cfc81e7ccbc49c88

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

• Loss: 1.0078
• Data Size: 1.0
• Epoch Runtime: 5.2641
• Accuracy: 0.7783
• F1 Macro: 0.7466
• Rouge1: 0.7777
• Rouge2: 0.0
• Rougel: 0.7789
• Rougelsum: 0.7789

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	0.7135	0	1.0609	0.3314	0.2496	0.3308	0.0	0.3317	0.3314
No log	1	114	0.6662	0.0078	1.3643	0.6651	0.3994	0.6657	0.0	0.6645	0.6651
No log	2	228	0.6408	0.0156	1.4235	0.6651	0.3994	0.6657	0.0	0.6645	0.6651
No log	3	342	0.6388	0.0312	1.6159	0.6651	0.3994	0.6657	0.0	0.6645	0.6651
0.0205	4	456	0.6150	0.0625	1.9014	0.6745	0.4306	0.6751	0.0	0.6745	0.6745
0.0205	5	570	0.6034	0.125	2.0959	0.6869	0.4818	0.6869	0.0	0.6869	0.6869
0.0205	6	684	0.5188	0.25	2.6063	0.7417	0.6524	0.7417	0.0	0.7417	0.7417
0.1345	7	798	0.4405	0.5	3.4198	0.8001	0.7643	0.7995	0.0	0.8001	0.8007
0.3533	8.0	912	0.4843	1.0	5.1439	0.7848	0.7230	0.7848	0.0	0.7854	0.7854
0.165	9.0	1026	0.6949	1.0	5.1065	0.7830	0.7197	0.7824	0.0	0.7830	0.7836
0.1128	10.0	1140	0.7659	1.0	5.1628	0.7978	0.7537	0.7983	0.0	0.7983	0.7981
0.0761	11.0	1254	1.0078	1.0	5.2641	0.7783	0.7466	0.7777	0.0	0.7789	0.7789

Framework versions

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