[BUG]: PlayGroud not extracting prompt for DSPY span

[RahulBalan]

Where do you use Phoenix

Local (self-hosted)

What version of Phoenix are you using?

8.8.0

What operating system are you seeing the problem on?

macOS

What version of Python are you running Phoenix with?

3.11.11

What version of Python or Node are you using instrumentation with?

No response

What instrumentation are you using?

No response

What happened?

We are trying to implement the experiment feature.
We use DSPY , not Openai chat completion directly.
We observe that, when we click the playground button on the GPT3.request span, it showing below error
"The following errors occurred when parsing span attributes:

Unable to parse span input messages, expected messages which include a role and content.
Unable to parse span output messages, expected messages which include a role and content."

What did you expect to happen?

we expect it to show the prompt just like how it works for openai chat completion

How can we reproduce the bug?

self explanatory

Additional information

No response

[cephalization]

Hi @RahulBalan, sorry to hear you are facing issues with this. Can you provide replication details such as dspy package version, instrumentation packages + versions, and possibly a code snippet?

[RahulBalan]

am using dspy.OpenAI.
But if i use dspy.LM , then playground is able to extract prompt.
so, this works only with dspy.LM ?