43a129ac79cba38c5df9c5da80286047

This model is a fine-tuned version of google-bert/bert-large-uncased-whole-word-masking on the contemmcm/clickbait dataset. It achieves the following results on the evaluation set:

• Loss: 0.6675
• Data Size: 1.0
• Epoch Runtime: 69.1137
• Accuracy: 0.6130
• F1 Macro: 0.3801

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	0.9545	0	4.6753	0.3866	0.2788
No log	1	650	0.5951	0.0078	5.7195	0.7135	0.6384
No log	2	1300	0.2596	0.0156	6.1922	0.9443	0.9425
No log	3	1950	0.0964	0.0312	8.5413	0.9774	0.9763
No log	4	2600	0.0491	0.0625	9.9262	0.9867	0.9860
0.0083	5	3250	0.0436	0.125	13.7153	0.9896	0.9890
0.0512	6	3900	0.0692	0.25	22.5037	0.9890	0.9884
0.0449	7	4550	0.0526	0.5	37.2871	0.9867	0.9859
0.6855	8.0	5200	0.6690	1.0	69.9261	0.6130	0.3801
0.6645	9.0	5850	0.6675	1.0	69.1137	0.6130	0.3801

Framework versions

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