c6b711ea53f6f00bab8e5b41903b18da

This model is a fine-tuned version of google/gemma-2b on the contemmcm/clickbait dataset. It achieves the following results on the evaluation set:

• Loss: 0.0818
• Data Size: 1.0
• Epoch Runtime: 125.0616
• Accuracy: 0.9981
• F1 Macro: 0.9980

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	3.6888	0	8.0392	0.4265	0.4026
No log	1	650	1.0954	0.0078	8.7097	0.9668	0.9644
No log	2	1300	0.0666	0.0156	13.6815	0.9983	0.9982
No log	3	1950	0.2233	0.0312	17.8762	0.9873	0.9866
No log	4	2600	0.2934	0.0625	23.1838	0.9904	0.9898
0.0167	5	3250	0.4719	0.125	32.7037	0.9817	0.9808
0.0741	6	3900	0.0354	0.25	49.5215	0.9990	0.9990
0.0959	7	4550	0.0875	0.5	81.9941	0.9977	0.9976
0.0001	8.0	5200	0.0641	1.0	130.0327	0.9983	0.9982
0.0001	9.0	5850	0.0514	1.0	123.6278	0.9988	0.9988
0.0059	10.0	6500	0.0818	1.0	125.0616	0.9981	0.9980

Framework versions

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