learning_rate: 3e-05
train_batch_size: 16
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 32
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 10
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
1.6857	1.0	115	0.4780	0.7966	0.8605
0.7405	2.0	230	0.4751	0.7892	0.8590
0.3852	3.0	345	0.9562	0.7892	0.8470
0.2413	4.0	460	1.3220	0.8186	0.8775
0.1375	5.0	575	1.4694	0.8235	0.8792
0.1035	6.0	690	1.6084	0.8186	0.8724
0.0603	7.0	805	1.8562	0.7966	0.8600
0.0172	8.0	920	1.6660	0.7892	0.8567
0.0112	9.0	1035	1.7459	0.7892	0.8571
0.006	10.0	1150	1.8153	0.7843	0.8543

jhj1769/Llama1B_Full_mrpc

Llama1B_Full_mrpc