Skip to content
By using AWS re:Post, you agree to the AWS re:Post Terms of Use
AWS re:Postre:Post
About re:Post

Location of code interpreter generated code

0

All, Here is the situation: I have a test agent with model Claude 3 Sonnet and simple S3 based KB. In my agent i enabled Code Interpreter in Advanced section. Agent provides a good response to the query. In response i see a link like </> - code interpreter. When i clink the link, i see message "Code was generated to enhance this response." and link to "show details". However, i do not see generated code. "Show details" link brings me to the trace. am i on a right path to see generated code? if i asked to generate some graph based on KB, i have a link to download resulting image, but no link to the code. Obviously i am missing something. Additional permissions? different model? Any information or working example would be greatly appreciated Thank you YK

Topics
Machine Learning & AIGenerative AI on AWS
Tags
Amazon Bedrock
Language
English
asked 2 years ago384 views
2 Answers
0

I would recommend reaching out to Anthropic's support team for more information and assistance regarding this feature. They should be able to provide you with the necessary details, guidance, and any additional permissions or configurations required to access and view the code generated by the Code Interpreter.

Additionally, it's worth noting that the availability and behavior of specific features like the Code Interpreter may vary depending on the model version, configurations, and the testing environment you're using. Anthropic's support team should be able to advise you on the appropriate setup and conditions required for this feature to work as expected.

EXPERT
answered 2 years ago
EXPERT
reviewed 2 years ago
  • YK
    2 years ago

    Thank you, i will try to find another model if Anthropic's models would not work.

0

I finally got some success... Using Anthropic model "Cloude 3 Sonnet" i created agent to advice on red light camera locations (my typical test case). KB was created based on S3 red camera location data. Simple prompt is "Using the red light camera data from KB create a bar graph of cameras on sheridan intersections. After generating the graph, provide a brief interpretation of the results".

As the result of this request i got a resulting graph. **i also got code - location is in trace (in my case i got Generated code for Trace Step 9) ** where i can see some Python code. In AWS CloudWatch trace file i can also see code and multiple mentioning of <tool_name>get::codeinterpreteraction::execute</tool_name>. All, please share your thoughts and suggestions. Thank you YK

answered 2 years ago

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.

Relevant content