distilbert_add_GLUE_Experiment_rte_192

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

• Loss: 0.6920
• Accuracy: 0.5271

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: 256
• eval_batch_size: 256
• seed: 10
• distributed_type: multi-GPU
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6952	1.0	10	0.6929	0.5271
0.6935	2.0	20	0.6924	0.5271
0.6944	3.0	30	0.6930	0.5271
0.6944	4.0	40	0.6930	0.5271
0.6944	5.0	50	0.6944	0.4729
0.6931	6.0	60	0.6921	0.5271
0.6942	7.0	70	0.6926	0.5271
0.6937	8.0	80	0.6939	0.4729
0.6934	9.0	90	0.6921	0.5271
0.694	10.0	100	0.6920	0.5271
0.6937	11.0	110	0.6945	0.4729
0.6934	12.0	120	0.6928	0.5271
0.6934	13.0	130	0.6924	0.5271
0.6934	14.0	140	0.6935	0.4729
0.6937	15.0	150	0.6944	0.4729

Framework versions

• Transformers 4.26.0
• Pytorch 1.14.0a0+410ce96
• Datasets 2.8.0
• Tokenizers 0.13.2