84ef776a521e711b5365ed0c84e36b2c

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

• Loss: 0.6934
• Data Size: 1.0
• Epoch Runtime: 10.0174
• Accuracy: 0.4219
• F1 Macro: 0.3178
• Rouge1: 0.4219
• Rouge2: 0.0
• Rougel: 0.4219
• Rougelsum: 0.4219

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	0.7006	0	0.8069	0.4375	0.3043	0.4375	0.0	0.4375	0.4375
No log	1	19	0.7107	0.0078	1.2751	0.4375	0.3043	0.4375	0.0	0.4375	0.4375
No log	2	38	0.7068	0.0156	2.0105	0.4375	0.3043	0.4375	0.0	0.4375	0.4375
No log	3	57	0.6970	0.0312	3.1303	0.4375	0.3043	0.4375	0.0	0.4375	0.4375
No log	4	76	0.6937	0.0625	4.4129	0.4688	0.4421	0.4688	0.0	0.4609	0.4688
No log	5	95	0.6987	0.125	5.9448	0.4375	0.3043	0.4375	0.0	0.4375	0.4375
0.0804	6	114	0.6992	0.25	6.6206	0.4375	0.3043	0.4375	0.0	0.4375	0.4375
0.0804	7	133	0.6928	0.5	8.4215	0.5625	0.36	0.5625	0.0	0.5625	0.5625
0.5208	8.0	152	0.6926	1.0	11.9210	0.5625	0.36	0.5625	0.0	0.5625	0.5625
0.5208	9.0	171	0.6870	1.0	11.4472	0.5625	0.36	0.5625	0.0	0.5625	0.5625
0.5208	10.0	190	0.6945	1.0	10.1243	0.4375	0.3043	0.4375	0.0	0.4375	0.4375
0.6993	11.0	209	0.6911	1.0	9.6496	0.5625	0.36	0.5625	0.0	0.5625	0.5625
0.6993	12.0	228	0.6891	1.0	10.2082	0.5625	0.36	0.5625	0.0	0.5625	0.5625
0.6993	13.0	247	0.6934	1.0	10.0174	0.4219	0.3178	0.4219	0.0	0.4219	0.4219

Framework versions

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