da9f71c7bd95ffef980859926c3fdb5a

This model is a fine-tuned version of studio-ousia/mluke-base on the nyu-mll/glue [cola] dataset. It achieves the following results on the evaluation set:

• Loss: 0.8121
• Data Size: 1.0
• Epoch Runtime: 28.8081
• Accuracy: 0.7754
• F1 Macro: 0.7167
• Rouge1: 0.7754
• Rouge2: 0.0
• Rougel: 0.7754
• Rougelsum: 0.7744

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	0.6947	0	1.6072	0.4834	0.4787	0.4824	0.0	0.4834	0.4824
No log	1	267	0.6964	0.0078	3.1576	0.6885	0.4078	0.6895	0.0	0.6885	0.6885
No log	2	534	0.6174	0.0156	3.3469	0.6885	0.4078	0.6895	0.0	0.6885	0.6885
No log	3	801	0.6024	0.0312	3.7069	0.6885	0.4078	0.6895	0.0	0.6885	0.6885
No log	4	1068	0.5837	0.0625	4.4404	0.7012	0.4679	0.7012	0.0	0.7012	0.7002
0.0352	5	1335	0.7137	0.125	6.2311	0.6895	0.4111	0.6904	0.0	0.6899	0.6895
0.5081	6	1602	0.5750	0.25	9.2259	0.7539	0.6377	0.7539	0.0	0.7539	0.7539
0.4648	7	1869	0.6338	0.5	15.3430	0.7324	0.5462	0.7324	0.0	0.7334	0.7324
0.4093	8.0	2136	0.5175	1.0	27.6485	0.7949	0.7273	0.7949	0.0	0.7949	0.7939
0.25	9.0	2403	0.6798	1.0	26.9089	0.7725	0.6739	0.7725	0.0	0.7725	0.7715
0.2317	10.0	2670	0.6264	1.0	27.4783	0.7764	0.6966	0.7754	0.0	0.7764	0.7754
0.1722	11.0	2937	0.8314	1.0	27.6669	0.7861	0.7122	0.7861	0.0	0.7861	0.7861
0.175	12.0	3204	0.8121	1.0	28.8081	0.7754	0.7167	0.7754	0.0	0.7754	0.7744

Framework versions

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