65a3e1a2e304a0b2e1118431fd0a5f60

This model is a fine-tuned version of distilbert/distilgpt2 on the contemmcm/clickbait dataset. It achieves the following results on the evaluation set:

• Loss: 0.0242
• Data Size: 1.0
• Epoch Runtime: 29.1527
• Accuracy: 0.9979
• F1 Macro: 0.9978

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
No log	0	0	5.0518	0	3.5783	0.3873	0.2797
No log	1	650	0.2177	0.0078	3.8926	0.9296	0.9264
No log	2	1300	0.0150	0.0156	3.9929	0.9967	0.9965
No log	3	1950	0.0046	0.0312	4.3827	0.9986	0.9986
No log	4	2600	0.0039	0.0625	5.0552	0.9988	0.9988
0.0064	5	3250	0.0032	0.125	6.7608	0.9990	0.9990
0.0034	6	3900	0.0124	0.25	9.9927	0.9983	0.9982
0.0002	7	4550	0.0107	0.5	15.8401	0.9985	0.9984
0.0012	8.0	5200	0.0041	1.0	29.0009	0.9992	0.9992
0.0	9.0	5850	0.0017	1.0	29.7472	0.9992	0.9992
0.0	10.0	6500	0.0047	1.0	28.6022	0.9994	0.9994
0.0	11.0	7150	0.0089	1.0	28.0580	0.9992	0.9992
0.0	12.0	7800	0.0057	1.0	28.4593	0.9992	0.9992
0.0041	13.0	8450	0.0242	1.0	29.1527	0.9979	0.9978

Framework versions

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