sst-bert-base-uncased

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

• Loss: 0.0201
• Mse: 0.0201

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: 5e-05
• train_batch_size: 16
• eval_batch_size: 32
• 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	Mse
0.0428	1.0	534	0.0218	0.0218
0.0169	2.0	1068	0.0216	0.0216
0.0096	3.0	1602	0.0231	0.0232
0.0066	4.0	2136	0.0222	0.0223
0.0046	5.0	2670	0.0220	0.0220
0.0035	6.0	3204	0.0209	0.0210
0.0029	7.0	3738	0.0226	0.0227
0.0025	8.0	4272	0.0211	0.0211
0.0023	9.0	4806	0.0207	0.0208
0.002	10.0	5340	0.0218	0.0218
0.0017	11.0	5874	0.0201	0.0202
0.0015	12.0	6408	0.0212	0.0212
0.0014	13.0	6942	0.0202	0.0202
0.0012	14.0	7476	0.0205	0.0206
0.0009	15.0	8010	0.0203	0.0203
0.0008	16.0	8544	0.0202	0.0202
0.0007	17.0	9078	0.0206	0.0207
0.0006	18.0	9612	0.0200	0.0200
0.0005	19.0	10146	0.0201	0.0201
0.0005	20.0	10680	0.0201	0.0201

Framework versions

• Transformers 4.37.0
• Pytorch 1.13.1+cu117
• Datasets 2.15.0
• Tokenizers 0.15.2