Update README.md

README.md

@@ -28,12 +28,11 @@ The model is trained for four epochs on the [crumb/flan-ul2-tinystories](https:/
 Training arguments:
 
 ```
-per_device_train_batch_size=32,
-gradient_accumulation_steps=4,
+per_device_train_batch_size=16,
+gradient_accumulation_steps=8,
 warmup_steps=128,
 num_train_epochs=4,
 learning_rate=2e-4,
-bf16=True,
 eval_steps=64,
 optim="adamw_torch",
 ```