About dario_bernardo

dario_bernardo

I have read parts of the Med-Gemini paper and I was intrigued by the uncertainty-guided search technique used by Med-Gemini models to improve the accuracy of their answers. In essence, It involves an iterative process of generating multiple reasoning...

dario_bernardo · 04-11-2024

Hello,I am encountering different LLM responses for the same prompt with the temperature set to 0.I am using gemini-1.0-pro-002, and I have noticed that, for some reason, setting the temperature to 0 does not always result in the LLM returning the sa...

dario_bernardo

pdf_file_url can be just a cloud storage location link, something starting with "gs://"

dario_bernardo

I ended up building one myself. It is just few lines of code anyway.sleep_attempts = 0sleep_time = 2while True: try: response = self.model.generate_content(contents) return response.text except ResourceExhausted as re: print(f"ResourceExhausted excep...

dario_bernardo

two things,first, you don't need to pass a `generation_config` object to both the constructor and the `generate_content` method. You can pass it either the constructor or the method.for example:```generation_config = { "temperature": 1, "max_output_t...

dario_bernardo

To fix this, you can either request a quota increase, or what I do is to wait some time and resubmit my request. For example I have wrote a small python script that every time get this exception wait few seconds and resubmit the request. You can adap...

dario_bernardo

I agree, I have also tried with temperature=0 and top_k=1, which in my understanding should return the same response across runs. But even with this settings the model returns different answers all he time. This behaviour was not present in earlier g...

Google Cloud Community

My Stats

dario_bernardo's Bio

Badges dario_bernardo Earned

Recent Activity

Feature Request: Uncertainty response score

Unexpected Behavior: Gemini-1.0-Pro-002 Returns Different Outputs at Temperature 0

Re: pdf analysis using gemini-1.5-pro

Re: How to implement exponential backoff with gemini? too many 504s

Re: why doesn't response_mime_type = json work here?

Re: Gemini Pro Quota Exceeded

Re: Unexpected Behavior: Gemini-1.0-Pro-002 Returns Different Outputs at Temperature 0