c8a3c5dcba70ded6236fbc73a23cab85

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

• Loss: 0.0085
• Data Size: 1.0
• Epoch Runtime: 37.5293
• Accuracy: 0.9988
• F1 Macro: 0.9988

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	0.6888	0	3.1410	0.5893	0.4811
No log	1	650	0.0670	0.0078	3.6137	0.9905	0.9900
No log	2	1300	0.0158	0.0156	4.1024	0.9969	0.9967
No log	3	1950	0.0088	0.0312	4.8092	0.9985	0.9984
No log	4	2600	0.0073	0.0625	5.8267	0.9985	0.9984
0.0019	5	3250	0.0049	0.125	8.1497	0.9988	0.9988
0.0004	6	3900	0.0051	0.25	12.5464	0.9988	0.9988
0.0089	7	4550	0.0130	0.5	21.4665	0.9983	0.9982
0.0054	8.0	5200	0.0119	1.0	40.3707	0.9985	0.9984
0.0096	9.0	5850	0.0085	1.0	37.5293	0.9988	0.9988

Framework versions

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