How can I remove/reduce the safety options?

[DimitrisAndreou]

Description of the bug:

My python code looks like this:

      response = await model.generate_content_async(
        prompt,
        generation_config={
          'temperature': 0.0, 'max_output_tokens': expectedOutputTokens
        },
        safety_settings={
            "HARM_CATEGORY_DANGEROUS": "block_none",
            "HARM_CATEGORY_HARASSMENT": "block_none",
            "HARM_CATEGORY_HATE_SPEECH": "block_none",
            "HARM_CATEGORY_SEXUAL": "block_none"
            })

It produces internal errors (500) when I try to specify safety_settings as above. Is this not the right way? I'm trying to reduce the number of errors I get due to safety. Thanks!

Actual vs expected behavior:

Would expect this either to fail early with a message telling me what's wrong with my arguments, or for it to work as expected, while producing fewer errors of the sort "response blocked because of safety"

Any other information you'd like to share?

No response