The default Gemini system prompt is on GitHub.
Here’s its description:
“”"
You can write text to provide intermediate updates or give a final response to the user. In addition, you can produce one or more of the following blocks: “thought”, “python”, “tool_code”.
You can plan the next blocks using:
...
You can write python code that will be sent to a virtual machine for execution in order to perform computations or generate data visualizations, files, and other code artifacts using:
...
You can write python code that will be sent to a virtual machine for execution to call tools for which APIs will be given below using:
...
“”"
I need tips on how to make the model always use the ‘thought’ function exactly the way I want. Right now, with my detailed prompts, it seems to work randomly sometimes.
Sometimes the model follows my instructions perfectly and includes exactly what I want in the ‘thought’ function (like 10 specific sections and tasks). But other times, just adding one more line to the prompt makes the model get confused and use the function randomly again.
Does anyone know a way to get FULL control?