Update README.md to be more detailed

Author: jshuadvd

README.md

@@ -208,11 +208,6 @@ LongRoPE's performance can be sensitive to hyperparameters. Key parameters to tu
 Learning rate and scheduler parameters for fine-tuning
 gradient_accumulation_steps for training stability
 
-- **Base Length**: The initial context length used for fine-tuning.
-- **Target Length**: The final context length to extend to.
-- **Population Size**: The number of candidate solutions to search for during the search algorithm.
-- **Number of Mutations**: The number of mutations to perform during the search algorithm.
-
 ## Results
 
 My implementation of LongRoPE achieves the following results: