Moderator-Chan_GPT-JT-6B: benchmark results
Provider: Other. Access: Open.
Aggregate Rating 1205 ± 15, rank #9587 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 - College Mathematics | 38 | Accuracy (%) | 89.5 |
| MMLU-by-task - Econometrics | 33.33 | Accuracy (%) | 80.1 |
| MMLU-by-task - Virology | 41.57 | Accuracy (%) | 53.6 |
| MMLU-by-task - Conceptual Physics | 39.15 | Accuracy (%) | 49.1 |
| MMLU-by-task - Machine Learning | 32.14 | Accuracy (%) | 48.9 |
| MMLU-by-task - High School Chemistry | 31.53 | Accuracy (%) | 45.2 |
| MMLU-by-task - Anatomy | 42.96 | Accuracy (%) | 45.1 |
| MMLU-by-task - Global Facts | 32 | Accuracy (%) | 44.3 |
| MMLU-by-task - Human Aging | 49.33 | Accuracy (%) | 43.5 |
| MMLU-by-task - College Medicine | 36.42 | Accuracy (%) | 43.2 |
| MMLU-by-task - Astronomy | 38.82 | Accuracy (%) | 41.9 |
Interactive version: theaggregate.ai/model?slug=moderator-chan-gpt-jt-6b · How It Works · Data refreshed daily, snapshot 2026-11-08.