Tracing requests sent to LLMs

[testy-cool]

Is there any way to check requests sent to LLMs so we can optimize prompts?

When using Burr it throws an error ValueError: Action Fetch attempted to write to keys {'original_html'} that it did not declare. It declared: (['doc'])!

[dosubot[bot]]

Hello @testy-cool! I'm Dosu, and I'm here to help you with bugs, answer your questions, and assist you in becoming a contributor. While you wait for a human maintainer, feel free to share more about your issue!

The error you're encountering with Burr, ValueError: Action Fetch attempted to write to keys {'original_html'} that it did not declare, is likely due to the HtmlAnalyzerNode expecting the original_html key to be present in the state. This key is required for the node to function correctly, as it needs access to the original HTML content for processing. Ensure that your state includes the original_html key when executing the node ^[1].

Regarding tracing requests sent to LLMs, I couldn't find specific information on how Burr logs or traces these requests. However, you might want to explore the configuration options in the graph_config to see if there are any logging parameters you can enable.

To continue talking to Dosu, mention @dosu.

---

^{Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Bug Report | Other}