2d28f2dd78d0fd0d608e00c42508795b

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

• Loss: 0.6572
• Data Size: 1.0
• Epoch Runtime: 1932.6824
• Accuracy: 0.6320
• F1 Macro: 0.3872
• Rouge1: 0.6318
• Rouge2: 0.0
• Rougel: 0.6319
• Rougelsum: 0.6317

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	0.7068	0	59.0808	0.3695	0.2739	0.3696	0.0	0.3695	0.3697
0.6935	1	11370	0.6587	0.0078	76.4379	0.6320	0.3872	0.6318	0.0	0.6319	0.6317
0.6502	2	22740	0.6579	0.0156	89.9177	0.6320	0.3872	0.6318	0.0	0.6319	0.6317
0.6533	3	34110	0.6598	0.0312	120.0155	0.6320	0.3872	0.6318	0.0	0.6319	0.6317
0.6615	4	45480	0.6568	0.0625	179.5245	0.6320	0.3872	0.6318	0.0	0.6319	0.6317
0.6598	5	56850	0.6587	0.125	296.0138	0.6320	0.3872	0.6318	0.0	0.6319	0.6317
0.6526	6	68220	0.6578	0.25	529.2549	0.6320	0.3872	0.6318	0.0	0.6319	0.6317
0.6546	7	79590	0.6583	0.5	995.4102	0.6320	0.3872	0.6318	0.0	0.6319	0.6317
0.6545	8.0	90960	0.6572	1.0	1932.6824	0.6320	0.3872	0.6318	0.0	0.6319	0.6317

Framework versions

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