Model save

README.md

@@ -36,14 +36,14 @@ The following hyperparameters were used during training:
 - eval_batch_size: 1
 - seed: 42
 - distributed_type: multi-GPU
-- num_devices: 96
+- num_devices: 104
 - gradient_accumulation_steps: 5
-- total_train_batch_size: 480
-- total_eval_batch_size: 96
+- total_train_batch_size: 520
+- total_eval_batch_size: 104
 - 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
 - lr_scheduler_warmup_steps: 100
-- num_epochs: 100
+- num_epochs: 300
 
 ### Training results