Modifying the Learning Rate in the middle of the Model Training in Deep Learning

Question

Below is the code to configure TrainingArguments consumed from the HuggingFace transformers library to finetune the GPT2 language model.

training_args = TrainingA