Update Hypers

README.md

@@ -180,15 +180,12 @@ Custom synthetic
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-04
-- train_batch_size: 1
-- eval_batch_size: 1
+- train_batch_size: 10
+- eval_batch_size: 3
 - distributed_type: multi-GPU
 - num_devices: 2
-- total_train_batch_size: 100
-- total_eval_batch_size: 10
 - optimizer: Adam 8bit
 - lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 10
 - num_epochs: 3
 
 ### Training results