asun17904/anliR1-t5-base

Knowledge Continuity Regularized Network

Dataset: ANLI Round: None

Trainer Hyperparameters:

• lr = 5e-05
• per_device_batch_size = 32
• gradient_accumulation_steps = 1
• weight_decay = 1e-09
• seed = 42

Regularization Hyperparameters

• numerical stability denominator constant = 1.0
• lambda = 1.0
• alpha = 1.0
• beta = 1.0

Extended Logs:

eval_loss	eval_accuracy	epoch
1.090	0.375	1.0
1.127	0.401	2.0
1.127	0.405	3.0
1.101	0.428	4.0
1.094	0.435	5.0
1.096	0.443	6.0
1.094	0.444	7.0
1.090	0.444	8.0
1.080	0.458	9.0
1.077	0.463	10.0
1.088	0.451	11.0
1.079	0.468	12.0
1.074	0.471	13.0
1.084	0.460	14.0
1.080	0.461	15.0
1.084	0.462	16.0
1.084	0.463	17.0
1.083	0.463	18.0
1.083	0.461	19.0

Test Accuracy: 0.331