Llama 2 16B Nastychat: benchmark results
Provider: chargoddard. Access: Open.
Aggregate Rating 1345 ± 15, rank #7831 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - Abstract Algebra | 37 | Accuracy (%) | 94.5 |
| MMLU-by-task - TruthfulQA MC2 | 53.45 | Accuracy (%) | 90.4 |
| MMLU-by-task - Electrical Engineering | 55.17 | Accuracy (%) | 89.9 |
| MMLU-by-task - International Law | 77.69 | Accuracy (%) | 88.6 |
| MMLU-by-task - College Computer Science | 50 | Accuracy (%) | 88.1 |
| MMLU-by-task - TruthfulQA MC1 | 36.47 | Accuracy (%) | 87.8 |
| MMLU-by-task - US Foreign Policy | 84 | Accuracy (%) | 86.5 |
| MMLU-by-task - Clinical Knowledge | 62.26 | Accuracy (%) | 86.3 |
| MMLU-by-task - Human Aging | 66.82 | Accuracy (%) | 84 |
| MMLU-by-task - Virology | 48.19 | Accuracy (%) | 83.9 |
| MMLU-by-task - Philosophy | 65.27 | Accuracy (%) | 82.9 |
| MMLU-by-task - Anatomy | 51.11 | Accuracy (%) | 82.4 |
Interactive version: theaggregate.ai/model?slug=llama-2-16b-nastychat · How It Works · Data refreshed daily, snapshot 2026-11-08.