bert-large-uncased-nsp-5000-1e-06-8

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: 0.3495

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: 32
• 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
No log	1.0	157	0.6936
0.7029	2.0	314	0.6880
0.6941	3.0	471	0.6812
0.6881	4.0	628	0.6655
0.6881	5.0	785	0.6302
0.6608	6.0	942	0.5869
0.6148	7.0	1099	0.5401
0.5525	8.0	1256	0.4810
0.4768	9.0	1413	0.4299
0.4768	10.0	1570	0.3978
0.4037	11.0	1727	0.3830
0.3647	12.0	1884	0.3699
0.318	13.0	2041	0.3633
0.318	14.0	2198	0.3557
0.2978	15.0	2355	0.3516
0.2774	16.0	2512	0.3529
0.2447	17.0	2669	0.3499
0.2435	18.0	2826	0.3495
0.2435	19.0	2983	0.3544
0.2231	20.0	3140	0.3517
0.2174	21.0	3297	0.3514

Framework versions

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