e5bacf463c7f03089bbd97817048989b

This model is a fine-tuned version of albert/albert-base-v2 on the contemmcm/clickbait dataset. It achieves the following results on the evaluation set:

• Loss: 0.0934
• Data Size: 1.0
• Epoch Runtime: 27.3014
• Accuracy: 0.9857
• F1 Macro: 0.9849

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	0.6673	0	2.5890	0.6319	0.6222
No log	1	650	0.2703	0.0078	3.0839	0.9273	0.9204
No log	2	1300	0.1012	0.0156	3.0732	0.9742	0.9728
No log	3	1950	0.1109	0.0312	3.4173	0.9635	0.9610
No log	4	2600	0.0767	0.0625	4.1559	0.9823	0.9813
0.0083	5	3250	0.1076	0.125	5.7580	0.9738	0.9726
0.1036	6	3900	0.3197	0.25	8.6107	0.9500	0.9461
0.1548	7	4550	0.0636	0.5	14.7882	0.9767	0.9752
0.0413	8.0	5200	0.0631	1.0	27.5855	0.9861	0.9853
0.072	9.0	5850	0.0661	1.0	27.9251	0.9869	0.9862
0.0494	10.0	6500	0.1063	1.0	27.9271	0.9774	0.9760
0.0177	11.0	7150	0.0686	1.0	27.2104	0.9853	0.9845
0.0266	12.0	7800	0.0934	1.0	27.3014	0.9857	0.9849

Framework versions

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