SOMD-bert-stage2-v1

This model is a fine-tuned version of bert-base-cased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0122

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: 16
• eval_batch_size: 8
• 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
No log	1.0	202	0.7166
No log	1.99	404	0.3823
0.8037	2.99	606	0.2156
0.8037	3.98	808	0.1470
0.2249	4.98	1010	0.0814
0.2249	5.97	1212	0.0568
0.2249	6.97	1414	0.0352
0.0842	7.96	1616	0.0320
0.0842	8.96	1818	0.0289
0.0512	9.95	2020	0.0255
0.0512	10.95	2222	0.0241
0.0512	11.94	2424	0.0228
0.0358	12.94	2626	0.0263
0.0358	13.93	2828	0.0149
0.0277	14.93	3030	0.0156
0.0277	15.92	3232	0.0147
0.0277	16.92	3434	0.0125
0.0224	17.91	3636	0.0140
0.0224	18.91	3838	0.0125
0.0188	19.9	4040	0.0122

Framework versions

• Transformers 4.38.1
• Pytorch 2.1.0+cu121
• Datasets 2.17.1
• Tokenizers 0.15.2