{% extends "base.html" %} {% block title %}Benchmark Results - lm-eval-ledger{% endblock %} {% block content %}
Benchmark results
| run | model | task (few-shot) | acc | n | no answer | time | tok/s |
|---|---|---|---|---|---|---|---|
| {{ b.run_id }} | {{ b.model_tag }} | {{ b.task }}{% if b.fewshot_k is not none %}({{ b.fewshot_k }}){% endif %} | {{ medals.get(b.benchmark_id, "") }}{% if b.verified_accuracy is not none %} {{ "%.3f"|format(b.verified_accuracy) }}✓ {% else %}{{ "%.3f"|format(b.accuracy) if b.accuracy is not none else "…" }}{% endif %} | {{ b.total_examples }} | {{ b.no_answer_count }} | {{ b.duration_seconds|fmtdur if b.duration_seconds else "-" }} | {{ "%.0f"|format(b.gen_tokens / b.gen_seconds) if b.gen_tokens and b.gen_seconds else "-" }} |