GPT-J-Pyg_PPO-6B-Dev-V8p4: benchmark results
Provider: EleutherAI. Access: Open.
Aggregate Rating 1168 ± 16, rank #9784 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - College Mathematics | 37 | Accuracy (%) | 85.2 |
| MMLU-by-task - Moral Scenarios | 27.26 | Accuracy (%) | 59.8 |
| MMLU-by-task - Global Facts | 34 | Accuracy (%) | 58.4 |
| MMLU-by-task - High School Physics | 31.13 | Accuracy (%) | 56.1 |
| MMLU-by-task - Econometrics | 28.07 | Accuracy (%) | 47.3 |
| MMLU-by-task - Astronomy | 38.82 | Accuracy (%) | 41.9 |
| MMLU-by-task - College Physics | 23.53 | Accuracy (%) | 41.6 |
| MMLU-by-task - Abstract Algebra | 28 | Accuracy (%) | 39.6 |
| MMLU-by-task - Electrical Engineering | 35.86 | Accuracy (%) | 39.6 |
| MMLU-by-task - High School Macroeconomics | 36.41 | Accuracy (%) | 39.5 |
| MMLU-by-task - College Chemistry | 29 | Accuracy (%) | 38 |
| MMLU-by-task - Professional Law | 29.73 | Accuracy (%) | 37.5 |
Interactive version: theaggregate.ai/model?slug=gpt-j-pyg-ppo-6b-dev-v8p4 · How It Works · Data refreshed daily, snapshot 2026-11-08.