bert-base-uncased-finetuned-CLS-RTE

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

• Loss: 2.2843
• Accuracy: 0.6968

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: 16
• 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	Accuracy
No log	1.0	156	0.6285	0.6390
No log	2.0	312	0.6483	0.6679
No log	3.0	468	0.8430	0.7004
0.4504	4.0	624	1.2886	0.6606
0.4504	5.0	780	1.6445	0.7004
0.4504	6.0	936	1.8099	0.6751
0.0631	7.0	1092	1.8173	0.7040
0.0631	8.0	1248	1.9799	0.7004
0.0631	9.0	1404	1.9767	0.6968
0.0159	10.0	1560	2.1696	0.6715
0.0159	11.0	1716	2.2142	0.6859
0.0159	12.0	1872	2.2605	0.6823
0.0076	13.0	2028	2.2775	0.7040
0.0076	14.0	2184	2.2719	0.6968
0.0076	15.0	2340	2.2843	0.6968

Framework versions

• Transformers 4.36.1
• Pytorch 2.1.2
• Datasets 2.15.0
• Tokenizers 0.15.0