distilbert-base-uncased-finetuned-clinc

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

• Loss: 0.0311
• Accuracy: 0.94

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: 1.9448451109337938e-05
• train_batch_size: 48
• eval_batch_size: 48
• 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
0.4446	1.0	318	0.2399	0.6155
0.1981	2.0	636	0.1226	0.8468
0.1246	3.0	954	0.0827	0.8987
0.0927	4.0	1272	0.0634	0.9158
0.0751	5.0	1590	0.0528	0.9232
0.0646	6.0	1908	0.0458	0.9313
0.0579	7.0	2226	0.0419	0.9316
0.053	8.0	2544	0.0398	0.9355
0.0497	9.0	2862	0.0368	0.9355
0.0471	10.0	3180	0.0357	0.9384
0.0449	11.0	3498	0.0349	0.9361
0.0434	12.0	3816	0.0340	0.9377
0.042	13.0	4134	0.0332	0.9371
0.0409	14.0	4452	0.0324	0.9381
0.0401	15.0	4770	0.0323	0.9387
0.0392	16.0	5088	0.0318	0.9387
0.0389	17.0	5406	0.0315	0.9394
0.0383	18.0	5724	0.0313	0.94
0.038	19.0	6042	0.0311	0.94
0.0377	20.0	6360	0.0311	0.94

Framework versions

• Transformers 4.44.2
• Pytorch 2.5.0+cu121
• Datasets 3.0.2
• Tokenizers 0.19.1