2c0f92e73ccda29903c43183fe0d6e1a

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

• Loss: 0.0111
• Data Size: 1.0
• Epoch Runtime: 20.2914
• Accuracy: 0.9983
• F1 Macro: 0.9982

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	0.7287	0	1.9963	0.3862	0.2788
No log	1	650	0.2340	0.0078	2.5171	0.9840	0.9830
No log	2	1300	0.1051	0.0156	2.5878	0.9591	0.9577
No log	3	1950	0.0108	0.0312	2.8768	0.9965	0.9963
No log	4	2600	0.0049	0.0625	3.3877	0.9992	0.9992
0.0023	5	3250	0.0100	0.125	4.6309	0.9977	0.9976
0.0002	6	3900	0.0023	0.25	6.8143	0.9996	0.9996
0.001	7	4550	0.0143	0.5	11.2185	0.9979	0.9978
0.0001	8.0	5200	0.0045	1.0	20.4988	0.9992	0.9992
0.0	9.0	5850	0.0090	1.0	20.3910	0.9990	0.9990
0.0	10.0	6500	0.0111	1.0	20.2914	0.9983	0.9982

Framework versions

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