Update README.md

README.md

@@ -67,7 +67,7 @@ You should:
 
 
 ### Message length control
-Inspired by the previously named "Roleplay" preset in SillyTavern, starting from this
+Inspired by the previously named "Roleplay" preset in SillyTavern, with this
 version of LimaRP it is possible to append a length modifier to the response instruction
 sequence, like this:
 
@@ -111,13 +111,13 @@ training process closer to a full finetune. It's suggested to merge the adapter
 the base Mistral-7B-v0.1 model.
 
 ### Training hyperparameters
-- learning_rate: 0.0010
+- learning_rate: 0.001
 - lr_scheduler_type: cosine
+- num_epochs: 2
 - lora_r: 256
 - lora_alpha: 16
 - lora_dropout: 0.05
 - lora_target_linear: True
-- num_epochs: 2
 - bf16: True
 - tf32: True
 - load_in_8bit: True