{% extends "base.html" %} {% block title %}{{ id }} — bench dashboard{% endblock %} {% block content %}

← {{ run_id }}

{{ id }} {{ outcome_label }}

Metadata

{% for row in meta_rows %} {% endfor %}
{{ row.label }}{{ row.value }}
{% match rubric %} {% when Some(r) %}

Triage mean {{ r.mean }}

{% if r.hacking_suspected %}

⚠ reward hacking suspected{% if !r.hacking_evidence.is_empty() %}: {{ r.hacking_evidence }}{% endif %}

{% endif %}

Verdict: {{ r.verdict }}

{% for s in r.scores %} {% endfor %}
rubricgradecomment
{{ s.label }}{{ s.grade }}{{ s.comment }}
{% if !r.observations.is_empty() %}

Observations

{% endif %}
{% when None %}

No rubric record for this rollout.

{% endmatch %}
{% if trajectory_note.is_empty() %}

Trajectory

{{ trajectory_html|safe }}
{% else %}

Trajectory

{{ trajectory_note }}

{% endif %}
{% endblock %}