GPT-J-Pyg_PPO-6B: benchmark results
Provider: EleutherAI. Access: Open.
Aggregate Rating 1161 ± 16, rank #9823 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - College Mathematics | 34 | Accuracy (%) | 69.2 |
| MMLU-by-task - Econometrics | 29.82 | Accuracy (%) | 62.1 |
| MMLU-by-task - Moral Scenarios | 27.15 | Accuracy (%) | 58.3 |
| MMLU-by-task - Global Facts | 32 | Accuracy (%) | 44.3 |
| MMLU-by-task - Security Studies | 42.86 | Accuracy (%) | 43.3 |
| MMLU-by-task - College Physics | 23.53 | Accuracy (%) | 41.6 |
| MMLU-by-task - Elementary Mathematics | 27.25 | Accuracy (%) | 41.4 |
| MMLU-by-task - Electrical Engineering | 33.79 | Accuracy (%) | 37.9 |
| MMLU-by-task - Astronomy | 34.21 | Accuracy (%) | 37.1 |
| MMLU-by-task - International Law | 43.8 | Accuracy (%) | 35.8 |
| MMLU-by-task - Professional Psychology | 30.88 | Accuracy (%) | 35.3 |
| MMLU-by-task - Sociology | 37.81 | Accuracy (%) | 35.2 |
Interactive version: theaggregate.ai/model?slug=gpt-j-pyg-ppo-6b · How It Works · Data refreshed daily, snapshot 2026-11-08.