621033441699ab5e5f3dac26085d0114

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

• Loss: 0.0796
• Data Size: 1.0
• Epoch Runtime: 19.4166
• Accuracy: 0.9878
• F1 Macro: 0.9872

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	0.6998	0	2.0935	0.3696	0.3043
No log	1	650	0.3791	0.0078	2.4622	0.9755	0.9742
No log	2	1300	0.0755	0.0156	2.5754	0.9784	0.9773
No log	3	1950	0.0596	0.0312	2.8019	0.9809	0.9799
No log	4	2600	0.0452	0.0625	3.4187	0.9880	0.9874
0.0047	5	3250	0.0356	0.125	4.3852	0.9905	0.9901
0.0257	6	3900	0.0764	0.25	6.7194	0.9805	0.9793
0.0198	7	4550	0.0273	0.5	10.8434	0.9921	0.9917
0.0267	8.0	5200	0.0270	1.0	19.6805	0.9931	0.9927
0.0031	9.0	5850	0.0459	1.0	19.7366	0.9936	0.9933
0.0	10.0	6500	0.0487	1.0	19.4059	0.9932	0.9929
0.0009	11.0	7150	0.0604	1.0	19.4612	0.9919	0.9915
0.0068	12.0	7800	0.0796	1.0	19.4166	0.9878	0.9872

Framework versions

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