a6b980ae51319b5e521892c609311045

This model is a fine-tuned version of albert/albert-base-v1 on the contemmcm/clickbait dataset. It achieves the following results on the evaluation set:

• Loss: 0.0720
• Data Size: 1.0
• Epoch Runtime: 23.7511
• Accuracy: 0.9888
• F1 Macro: 0.9882

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	0.7299	0	2.2974	0.3463	0.3357
No log	1	650	0.2785	0.0078	2.7538	0.9186	0.9146
No log	2	1300	0.1167	0.0156	2.6886	0.9622	0.9602
No log	3	1950	0.0947	0.0312	3.1118	0.9713	0.9696
No log	4	2600	0.1628	0.0625	3.7764	0.9605	0.9575
0.0061	5	3250	0.0526	0.125	5.0750	0.9861	0.9854
0.0348	6	3900	0.0662	0.25	7.8565	0.9863	0.9856
0.0484	7	4550	0.0393	0.5	13.1783	0.9880	0.9874
0.0305	8.0	5200	0.0486	1.0	24.0609	0.9911	0.9907
0.0273	9.0	5850	0.0377	1.0	24.9323	0.9905	0.9900
0.0117	10.0	6500	0.0432	1.0	23.5958	0.9915	0.9911
0.0106	11.0	7150	0.0519	1.0	23.3127	0.9923	0.9919
0.0185	12.0	7800	0.0492	1.0	23.8838	0.9913	0.9909
0.0064	13.0	8450	0.0720	1.0	23.7511	0.9888	0.9882

Framework versions

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