1634bf48a582aeb4249707a56e0a282b

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

• Loss: 0.0123
• Data Size: 1.0
• Epoch Runtime: 34.6905
• Accuracy: 0.9985
• F1 Macro: 0.9984

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	0.6904	0	3.2978	0.5129	0.5118
No log	1	650	0.0340	0.0078	3.7297	0.9956	0.9953
No log	2	1300	0.0221	0.0156	3.8633	0.9963	0.9961
No log	3	1950	0.0343	0.0312	4.1301	0.9952	0.9949
No log	4	2600	0.0190	0.0625	5.1533	0.9973	0.9972
0.0014	5	3250	0.0127	0.125	7.1044	0.9977	0.9976
0.0111	6	3900	0.0206	0.25	10.8933	0.9971	0.9969
0.0341	7	4550	0.0096	0.5	18.2094	0.9977	0.9976
0.1621	8.0	5200	0.0159	1.0	35.8475	0.9979	0.9978
0.0065	9.0	5850	0.0241	1.0	34.2765	0.9973	0.9972
0.0021	10.0	6500	0.0135	1.0	35.7171	0.9979	0.9978
0.0087	11.0	7150	0.0123	1.0	34.6905	0.9985	0.9984

Framework versions

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