distilbert-base-uncased-clickbait-task1-20-epoch-post

This model is a fine-tuned version of distilbert/distilbert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.1587
• Accuracy: 0.6975

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	200	0.7981	0.67
No log	2.0	400	0.7490	0.695
0.7714	3.0	600	0.7823	0.7125
0.7714	4.0	800	0.8867	0.7025
0.3422	5.0	1000	1.0104	0.69
0.3422	6.0	1200	1.2018	0.685
0.3422	7.0	1400	1.4174	0.71
0.1036	8.0	1600	1.6449	0.7025
0.1036	9.0	1800	1.8061	0.6925
0.0419	10.0	2000	1.7564	0.71
0.0419	11.0	2200	1.8920	0.695
0.0419	12.0	2400	1.9388	0.6875
0.0174	13.0	2600	2.0049	0.695
0.0174	14.0	2800	2.1282	0.695
0.0116	15.0	3000	2.0698	0.6875
0.0116	16.0	3200	2.1423	0.7025
0.0116	17.0	3400	2.1413	0.69
0.0059	18.0	3600	2.1745	0.69
0.0059	19.0	3800	2.1527	0.695
0.0066	20.0	4000	2.1587	0.6975

Framework versions

• Transformers 4.44.0.dev0
• Pytorch 2.4.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1