b3ff764e061a7f5ed8ba076715d53a60

This model is a fine-tuned version of openai-community/gpt2-medium on the nyu-mll/glue [cola] dataset. It achieves the following results on the evaluation set:

• Loss: 1.2022
• Data Size: 1.0
• Epoch Runtime: 39.3126
• Accuracy: 0.7783
• F1 Macro: 0.6985
• Rouge1: 0.7783
• Rouge2: 0.0
• Rougel: 0.7773
• Rougelsum: 0.7783

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	6.9971	0	2.2560	0.6885	0.4108	0.6895	0.0	0.6885	0.6885
No log	1	267	1.3931	0.0078	2.6884	0.6006	0.5402	0.6016	0.0	0.6006	0.6006
No log	2	534	0.8726	0.0156	3.2968	0.6328	0.5468	0.6328	0.0	0.6318	0.6328
No log	3	801	0.9646	0.0312	4.3321	0.6885	0.4137	0.6895	0.0	0.6895	0.6895
No log	4	1068	0.6509	0.0625	5.7832	0.6924	0.4211	0.6934	0.0	0.6924	0.6924
0.0558	5	1335	0.9177	0.125	8.5274	0.6865	0.4071	0.6875	0.0	0.6865	0.6865
0.5915	6	1602	0.8271	0.25	13.1586	0.6982	0.4433	0.6982	0.0	0.6982	0.6982
0.5018	7	1869	0.7042	0.5	22.1718	0.7305	0.5541	0.7310	0.0	0.7314	0.7305
0.407	8.0	2136	0.5059	1.0	39.3019	0.7852	0.7089	0.7842	0.0	0.7852	0.7847
0.248	9.0	2403	0.7009	1.0	38.6571	0.7832	0.6916	0.7832	0.0	0.7832	0.7832
0.199	10.0	2670	0.7691	1.0	40.4511	0.7783	0.7114	0.7783	0.0	0.7783	0.7783
0.1354	11.0	2937	0.8919	1.0	39.4557	0.7822	0.7158	0.7812	0.0	0.7822	0.7822
0.125	12.0	3204	1.2022	1.0	39.3126	0.7783	0.6985	0.7783	0.0	0.7773	0.7783

Framework versions

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