4ee0584476aedf04614c90d837241a82

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

• Loss: 0.7337
• Data Size: 1.0
• Epoch Runtime: 20.7250
• Accuracy: 0.7646
• F1 Macro: 0.6966
• Rouge1: 0.7646
• Rouge2: 0.0
• Rougel: 0.7646
• Rougelsum: 0.7651

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	10.9000	0	1.6733	0.3115	0.2375	0.3105	0.0	0.3115	0.3115
No log	1	267	3.8392	0.0078	1.7776	0.3096	0.2364	0.3086	0.0	0.3086	0.3096
No log	2	534	1.9189	0.0156	1.9111	0.6758	0.4356	0.6768	0.0	0.6758	0.6748
No log	3	801	1.0215	0.0312	2.3998	0.3633	0.3444	0.3623	0.0	0.3628	0.3633
No log	4	1068	0.6360	0.0625	2.9400	0.6748	0.4519	0.6758	0.0	0.6748	0.6743
0.0742	5	1335	0.7139	0.125	4.3171	0.6875	0.4074	0.6885	0.0	0.6875	0.6875
0.5993	6	1602	0.6202	0.25	6.5947	0.6934	0.4356	0.6943	0.0	0.6934	0.6934
0.5756	7	1869	0.6145	0.5	11.4144	0.7041	0.4670	0.7051	0.0	0.7041	0.7051
0.4946	8.0	2136	0.5537	1.0	20.2694	0.7520	0.6312	0.7520	0.0	0.7520	0.7520
0.3546	9.0	2403	0.7079	1.0	19.8864	0.7588	0.6595	0.7588	0.0	0.7598	0.7588
0.3184	10.0	2670	0.6050	1.0	20.8954	0.7627	0.6807	0.7627	0.0	0.7627	0.7627
0.215	11.0	2937	0.6989	1.0	21.8290	0.7715	0.7078	0.7715	0.0	0.7715	0.7715
0.2029	12.0	3204	0.7337	1.0	20.7250	0.7646	0.6966	0.7646	0.0	0.7646	0.7651

Framework versions

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