6b6718d06cde06bca5b409f20a4aaf15

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

• Loss: 0.6805
• Data Size: 1.0
• Epoch Runtime: 67.1103
• 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	1.1075	0	4.6706	0.3870	0.2790
No log	1	650	0.6492	0.0078	5.5231	0.5336	0.5057
No log	2	1300	0.1599	0.0156	6.6493	0.9468	0.9450
No log	3	1950	0.1196	0.0312	8.2836	0.9603	0.9573
No log	4	2600	0.0484	0.0625	9.8174	0.9878	0.9872
0.0093	5	3250	0.0469	0.125	13.6634	0.9865	0.9858
0.0827	6	3900	0.0612	0.25	22.0063	0.9882	0.9876
0.082	7	4550	0.0460	0.5	36.3592	0.9888	0.9882
0.6843	8.0	5200	0.6704	1.0	67.4669	0.6130	0.3801
0.6642	9.0	5850	0.6686	1.0	66.8352	0.6130	0.3801
0.6643	10.0	6500	0.6693	1.0	66.0511	0.6130	0.3801
0.6852	11.0	7150	0.6805	1.0	67.1103	0.6130	0.3801

Framework versions

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