distilbert-base-uncased-distilled-clinc

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

• Loss: 0.3063
• Accuracy: 0.9494

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: 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: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	318	1.8614	0.7329
2.3518	2.0	636	0.7694	0.8565
2.3518	3.0	954	0.4345	0.9126
0.7002	4.0	1272	0.3663	0.9371
0.3298	5.0	1590	0.3418	0.9471
0.3298	6.0	1908	0.3318	0.9455
0.2692	7.0	2226	0.3242	0.9481
0.248	8.0	2544	0.3175	0.95
0.248	9.0	2862	0.3144	0.9494
0.2394	10.0	3180	0.3109	0.9513
0.2394	11.0	3498	0.3109	0.9497
0.2333	12.0	3816	0.3090	0.95
0.2305	13.0	4134	0.3089	0.9484
0.2305	14.0	4452	0.3074	0.9484
0.2278	15.0	4770	0.3063	0.9494

Framework versions

• Transformers 4.16.2
• Pytorch 2.4.1+cu121
• Datasets 1.16.1
• Tokenizers 0.19.1