docs: teach how to inspect each eval run in the Foundry portal#296
Merged
Conversation
The tutorial previously stopped at the terminal's aggregate pass/fail output, which makes the eval steps feel like a black box. Add a reusable `See the run in the Foundry portal` section at the end of step 10 (deep-link via results report_url + manual nav + what each pane shows) and link back to it from the step 11 multi-turn re-run and the step 11.4 rubric re-run. Step 11.4 also gets an explicit walkthrough of the `View rubric details` modal so users see the per-dimension judge scores even though the CLI threshold lives on the aggregate. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Adds a reusable section at the end of step 10 explaining how to open the run in https://ai.azure.com (deep link via report_url plus manual nav, with a description of what each pane on the run page shows). Links back to that section from the multi-turn re-run in step 11 and the rubric re-run in step 11.4. Step 11.4 also walks through the View rubric details modal so users see the per-dimension judge scores (correct_itinerary, clear_practical_notes, ...) even though the CLI threshold lives on the smoke-core aggregate.