e5949f8e378cc703b2a4e1e643b489fd

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

• Loss: 0.0064
• Data Size: 1.0
• Epoch Runtime: 30.2150
• Accuracy: 0.9992
• F1 Macro: 0.9992

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	0.7364	0	2.8061	0.3870	0.2790
No log	1	650	0.4304	0.0078	3.2226	0.7710	0.7112
No log	2	1300	0.0331	0.0156	3.4124	0.9919	0.9915
No log	3	1950	0.0086	0.0312	3.8829	0.9983	0.9982
No log	4	2600	0.0035	0.0625	4.8824	0.9992	0.9992
0.0031	5	3250	0.0022	0.125	6.5243	0.9994	0.9994
0.0028	6	3900	0.0044	0.25	9.7271	0.9990	0.9990
0.0027	7	4550	0.0059	0.5	16.4672	0.9988	0.9988
0.0018	8.0	5200	0.0013	1.0	30.0719	0.9994	0.9994
0.0146	9.0	5850	0.0126	1.0	29.0922	0.9985	0.9984
0.0	10.0	6500	0.0078	1.0	28.8978	0.9992	0.9992
0.0	11.0	7150	0.0054	1.0	29.1435	0.9992	0.9992
0.0	12.0	7800	0.0064	1.0	30.2150	0.9992	0.9992

Framework versions

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