xlm-roberta-base-clickbait-task1-20-epoch-post_title

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

• Loss: 2.6481
• Accuracy: 0.6875

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	0.8494	0.6675
No log	2.0	400	0.7488	0.6975
0.8603	3.0	600	0.7566	0.7025
0.8603	4.0	800	0.8778	0.7075
0.5018	5.0	1000	0.9345	0.7025
0.5018	6.0	1200	1.1088	0.695
0.5018	7.0	1400	1.4765	0.69
0.2227	8.0	1600	1.7443	0.6975
0.2227	9.0	1800	1.7676	0.6875
0.1063	10.0	2000	2.1244	0.675
0.1063	11.0	2200	2.2377	0.6925
0.1063	12.0	2400	2.3395	0.685
0.05	13.0	2600	2.4198	0.6925
0.05	14.0	2800	2.5012	0.6875
0.0218	15.0	3000	2.6027	0.675
0.0218	16.0	3200	2.5422	0.6925
0.0218	17.0	3400	2.6134	0.68
0.0138	18.0	3600	2.5818	0.69
0.0138	19.0	3800	2.6337	0.6825
0.0078	20.0	4000	2.6481	0.6875

Framework versions

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