This interactive explorer needs its page script, and it did not run. If you are reading this inside the blog, the explorer was embedded in a way that stripped its script: embed ta2_explorer.html as an iframe instead of inlining its markup.
Click a run or its name to select it; its details open below.
Double-click a run, its name, or the golden line to zoom to it alone. Double-click empty space or press All runs to return.
Click any step dot for that step's full judgment, golden steps included.
Scroll to zoom and drag to pan while zoomed.
Compare shows Opus 4.8 and GPT 5.6 Sol side by side.
Hover an overlay for a one-line explanation.
SteprelevantunclearirrelevantRun endpassedfailed* tokens estimated from run cost