pythia-13B-deduped-green_devil: benchmark results
Provider: EleutherAI.
Aggregate Rating 1144 ± 16, rank #10042 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - College Mathematics | 33 | Accuracy (%) | 61.7 |
| MMLU-by-task - Econometrics | 28.95 | Accuracy (%) | 55.2 |
| MMLU-by-task - Formal Logic | 30.95 | Accuracy (%) | 52.2 |
| MMLU-by-task - Elementary Mathematics | 28.84 | Accuracy (%) | 50.7 |
| MMLU-by-task - Moral Scenarios | 25.03 | Accuracy (%) | 44.1 |
| MMLU-by-task - High School Physics | 27.81 | Accuracy (%) | 35 |
| MMLU-by-task - College Physics | 22.55 | Accuracy (%) | 34.3 |
| MMLU-by-task - HellaSwag | 51.57 | Accuracy (%) | 33.5 |
| MMLU-by-task - World Religions | 34.5 | Accuracy (%) | 33.2 |
| MMLU-by-task - ARC Challenge | 38.74 | Accuracy (%) | 32.6 |
| MMLU-by-task - High School US History | 28.92 | Accuracy (%) | 32 |
| MMLU-by-task - Machine Learning | 28.57 | Accuracy (%) | 30.8 |
Interactive version: theaggregate.ai/model?slug=pythia-13b-deduped-green-devil · How It Works · Data refreshed daily, snapshot 2026-11-08.