835d0d4b30b544717551d2157dacc086

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

• Loss: 0.6592
• Data Size: 0.5
• Epoch Runtime: 549.1670
• 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	0.6614	0	31.5162	0.6306	0.3891	0.6305	0.0	0.6306	0.6304
0.6158	1	11370	0.6559	0.0078	41.8356	0.6384	0.4122	0.6383	0.0	0.6384	0.6382
0.6137	2	22740	0.4898	0.0156	48.5810	0.7706	0.7597	0.7706	0.0	0.7706	0.7706
0.547	3	34110	0.4836	0.0312	65.6771	0.7594	0.7412	0.7593	0.0	0.7594	0.7592
0.6798	4	45480	0.6603	0.0625	97.0398	0.6320	0.3872	0.6318	0.0	0.6319	0.6317
0.6703	5	56850	0.6586	0.125	160.7261	0.6320	0.3872	0.6318	0.0	0.6319	0.6317
0.6648	6	68220	0.6581	0.25	290.9691	0.6320	0.3872	0.6318	0.0	0.6319	0.6317
0.6585	7	79590	0.6592	0.5	549.1670	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