llama-13B-pretrained-sft-do2: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1294 ± 15, rank #8826 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - ARC Challenge | 56.48 | Accuracy (%) | 76.9 |
| MMLU-by-task - Anatomy | 49.63 | Accuracy (%) | 75.9 |
| MMLU-by-task - College Mathematics | 35 | Accuracy (%) | 75 |
| MMLU-by-task - TruthfulQA MC1 | 33.17 | Accuracy (%) | 71.8 |
| MMLU-by-task - College Physics | 29.41 | Accuracy (%) | 71.1 |
| MMLU-by-task - Moral Scenarios | 32.96 | Accuracy (%) | 70.8 |
| MMLU-by-task - HellaSwag | 60.81 | Accuracy (%) | 63 |
| MMLU-by-task - Computer Security | 64 | Accuracy (%) | 62.9 |
| MMLU-by-task - Elementary Mathematics | 30.95 | Accuracy (%) | 61.7 |
| MMLU-by-task - TruthfulQA MC2 | 47.41 | Accuracy (%) | 61.7 |
| MMLU-by-task - Marketing | 75.64 | Accuracy (%) | 61.3 |
| MMLU-by-task - High School Computer Science | 50 | Accuracy (%) | 60.7 |
Interactive version: theaggregate.ai/model?slug=llama-13b-pretrained-sft-do2 · How It Works · Data refreshed daily, snapshot 2026-11-08.