I am still thinking it might be an issue with the frontend (in your browser) contacting the backend (in kubernetes) however, and errors there may not propagate to the backend
I don't think the env var that runLLM suggested will help here, let me see what we have available. If you check the browser debug network tab when the error occurs you can see the full error in the meantime
how are you hosting phoenix? This reads to me like the application attempted to perform a graphql query and got plaintext back as a response, possibly from a proxy or similar
my recommendation would be to enter some filter on the data using the search bar (on metadata perhaps) and then select-all of the results and assign them to a split