llama-13B-pretrained-sft-epoch-1: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1294 ± 16, rank #8831 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - Global Facts | 37 | Accuracy (%) | 76.3 |
| MMLU-by-task - Electrical Engineering | 48.28 | Accuracy (%) | 68.5 |
| MMLU-by-task - ARC Challenge | 54.61 | Accuracy (%) | 67.2 |
| MMLU-by-task - US Foreign Policy | 76 | Accuracy (%) | 63.7 |
| MMLU-by-task - World Religions | 71.35 | Accuracy (%) | 60.5 |
| MMLU-by-task - Miscellaneous | 68.71 | Accuracy (%) | 60.1 |
| MMLU-by-task - TruthfulQA MC1 | 30.35 | Accuracy (%) | 59.4 |
| MMLU-by-task - Professional Law | 38.2 | Accuracy (%) | 59.3 |
| MMLU-by-task - Anatomy | 46.67 | Accuracy (%) | 57.5 |
| Open LLM Leaderboard v1 - WinoGrande | 77.58 | Accuracy (%) (5-shot) | 57 |
| MMLU-by-task - Professional Psychology | 48.04 | Accuracy (%) | 56.4 |
| MMLU-by-task - Management | 65.05 | Accuracy (%) | 56.3 |
Interactive version: theaggregate.ai/model?slug=llama-13b-pretrained-sft-epoch-1 · How It Works · Data refreshed daily, snapshot 2026-11-08.