fix(tracing): show model-call reasoning above the answer, not below#154
Open
arelchan wants to merge 1 commit into
Open
fix(tracing): show model-call reasoning above the answer, not below#154arelchan wants to merge 1 commit into
arelchan wants to merge 1 commit into
Conversation
The viewer's Model Output card appended the (folded) reasoning block after the assistant answer, so it read bottom-up. Reasoning precedes the answer chronologically -- the model thinks, then responds -- so render it at the top of the card, kept collapsed since it is long and diagnostic. Co-authored-by: Claude (claude-opus-4-8) <noreply@anthropic.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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Fix the reading order of a model-call detail in the tracing viewer: reasoning
now renders above the assistant answer instead of below it.
The Model Output card appended the (collapsed)
Reasoningblock after theanswer body, so the card read bottom-up. Reasoning precedes the answer
chronologically -- the model thinks, then responds -- so it belongs at the top
of the card. It stays collapsed by default since it is long and diagnostic.
Reasoning is part of the model's output (it comes from
llm.output'sreasoning_content), so it stays inside the Model Output card rather thanmoving to the Model Input area (system prompt / tools / history).
Card order is now:
Model Outputheader ->Reasoning(folded) -> answer.Type
Verification
Application changes have been tested thoroughly
node --check raven/tracing/viewer/ui/app.js-> OK.Booted the viewer against real traces and opened a
minimax/minimax-m3model call that carries reasoning. Confirmed the Model Output card children
are now
header -> Reasoning(details) -> answer(DOM checkreasoningBeforeAnswer: true) and verified it visually.Risk
Viewer-only, presentation-only change: no data model, schema, or API change;
purely reorders two blocks within one detail card.
Related Issues
N/A -- follow-up polish on the tracing viewer merged in #135.