f0a533efe507426ebb7be2b8405fc5e9

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

• Loss: 0.8987
• Data Size: 1.0
• Epoch Runtime: 39.3464
• Accuracy: 0.6117
• F1 Macro: 0.3795

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	0.6698	0	2.9391	0.6140	0.3828
No log	1	650	0.3231	0.0078	3.4729	0.9518	0.9484
No log	2	1300	0.2222	0.0156	4.0441	0.9188	0.9169
No log	3	1950	0.0994	0.0312	4.8344	0.9703	0.9685
No log	4	2600	0.1540	0.0625	6.0641	0.9574	0.9542
0.0078	5	3250	0.0606	0.125	8.1727	0.9842	0.9833
0.0609	6	3900	0.0821	0.25	12.8552	0.9826	0.9816
0.0392	7	4550	0.0581	0.5	21.6937	0.9871	0.9863
0.0226	8.0	5200	0.0401	1.0	40.3111	0.9911	0.9906
0.0235	9.0	5850	0.0635	1.0	38.5939	0.9886	0.9880
0.0112	10.0	6500	0.0614	1.0	39.5879	0.9900	0.9894
0.0179	11.0	7150	0.0814	1.0	38.1253	0.9875	0.9868
0.1752	12.0	7800	0.8987	1.0	39.3464	0.6117	0.3795

Framework versions

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