squad_distilbert

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: 2.5411

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: 32
• eval_batch_size: 32
• 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
1.2592	1.0	2738	1.1407
0.9254	2.0	5476	1.1297
0.7153	3.0	8214	1.1560
0.5814	4.0	10952	1.2990
0.4528	5.0	13690	1.4225
0.3592	6.0	16428	1.5411
0.2932	7.0	19166	1.6802
0.2383	8.0	21904	1.7984
0.1898	9.0	24642	1.9599
0.1553	10.0	27380	2.1424
0.1297	11.0	30118	2.2808
0.1074	12.0	32856	2.3198
0.0922	13.0	35594	2.3774
0.08	14.0	38332	2.5194
0.0678	15.0	41070	2.5411

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0