6aa4005550be4fa303acb85b50ebcc4b

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

• Loss: 0.5508
• Data Size: 1.0
• Epoch Runtime: 28.0586
• Accuracy: 0.8066
• F1 Macro: 0.7666
• Rouge1: 0.8066
• Rouge2: 0.0
• Rougel: 0.8066
• Rougelsum: 0.8076

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	0.6365	0	1.3498	0.6768	0.4310	0.6768	0.0	0.6768	0.6768
No log	1	267	0.6186	0.0078	2.7708	0.6943	0.4388	0.6943	0.0	0.6953	0.6943
No log	2	534	0.6401	0.0156	2.3427	0.6885	0.4078	0.6895	0.0	0.6885	0.6885
No log	3	801	0.5725	0.0312	2.9995	0.7461	0.6560	0.7461	0.0	0.7461	0.7461
No log	4	1068	0.5191	0.0625	4.4982	0.7715	0.6759	0.7725	0.0	0.7715	0.7705
0.0319	5	1335	0.7668	0.125	6.4782	0.6943	0.4360	0.6943	0.0	0.6943	0.6943
0.4635	6	1602	0.4476	0.25	10.2303	0.8086	0.7662	0.8086	0.0	0.8096	0.8086
0.4283	7	1869	0.4742	0.5	15.3332	0.8203	0.7686	0.8203	0.0	0.8203	0.8193
0.3991	8.0	2136	0.5609	1.0	28.4207	0.7148	0.5731	0.7148	0.0	0.7158	0.7148
0.3242	9.0	2403	0.5356	1.0	27.0623	0.8105	0.7752	0.8110	0.0	0.8096	0.8105
0.3066	10.0	2670	0.5508	1.0	28.0586	0.8066	0.7666	0.8066	0.0	0.8066	0.8076

Framework versions

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