bert-large-uncased-nsp-20000-1e-06-16

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

• Loss: 0.2694

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: 1e-06
• train_batch_size: 64
• eval_batch_size: 1024
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss
0.704	1.0	313	0.6745
0.6511	2.0	626	0.6037
0.5975	3.0	939	0.4653
0.4317	4.0	1252	0.3733
0.3829	5.0	1565	0.3359
0.3177	6.0	1878	0.3113
0.3097	7.0	2191	0.2955
0.2634	8.0	2504	0.2838
0.2392	9.0	2817	0.2795
0.2262	10.0	3130	0.2730
0.2051	11.0	3443	0.2714
0.1959	12.0	3756	0.2713
0.1852	13.0	4069	0.2694
0.1694	14.0	4382	0.2719
0.1627	15.0	4695	0.2713
0.1492	16.0	5008	0.2760

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1