549dbe460a2caadd404fc36330d5ba6e

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

• Loss: 0.6703
• Data Size: 1.0
• Epoch Runtime: 64.9198
• 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	0.9427	0	4.7000	0.3870	0.2790
No log	1	650	0.0546	0.0078	5.4242	0.9932	0.9929
No log	2	1300	0.0391	0.0156	6.2802	0.9931	0.9927
No log	3	1950	0.1306	0.0312	7.8490	0.9853	0.9844
No log	4	2600	0.0112	0.0625	9.7780	0.9977	0.9976
0.0027	5	3250	0.0073	0.125	13.3028	0.9981	0.9980
0.0072	6	3900	0.0233	0.25	21.8313	0.9961	0.9959
0.0159	7	4550	0.0887	0.5	36.0550	0.9767	0.9757
0.6816	8.0	5200	0.6686	1.0	67.3607	0.6130	0.3801
0.6738	9.0	5850	0.6703	1.0	64.9198	0.6130	0.3801

Framework versions

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