diff --git a/README.md b/README.md index 65efb12..274638d 100644 --- a/README.md +++ b/README.md @@ -206,7 +206,7 @@ LongRoPE's performance can be sensitive to hyperparameters. Key parameters to tu `population_size`, `num_mutations`, and `num_crossovers` in the lambda factor search Learning rate and scheduler parameters for fine-tuning -gradient_accumulation_steps for training stability +`gradient_accumulation_steps` for training stability ## Results