Similar pyspark logic returns different number of rows in dataframe

Question

Why do the following two lines of code produce a different result?

email_response.filter(f"first_response_date > date\'2020-11-2\'")