pythia-12B-sft-v8-2.5k-steps: benchmark results
Provider: EleutherAI. Access: Open.
Aggregate Rating 1154 ± 16, rank #9897 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - Abstract Algebra | 35 | Accuracy (%) | 87.4 |
| MMLU-by-task - High School Mathematics | 28.89 | Accuracy (%) | 67.8 |
| MMLU-by-task - College Physics | 23.53 | Accuracy (%) | 41.6 |
| MMLU-by-task - Econometrics | 27.19 | Accuracy (%) | 39.6 |
| MMLU-by-task - Conceptual Physics | 34.89 | Accuracy (%) | 38.5 |
| MMLU-by-task - Medical Genetics | 39 | Accuracy (%) | 36.9 |
| MMLU-by-task - High School Statistics | 32.41 | Accuracy (%) | 36.7 |
| MMLU-by-task - HellaSwag | 51.72 | Accuracy (%) | 33.8 |
| MMLU-by-task - Jurisprudence | 32.41 | Accuracy (%) | 33.7 |
| MMLU-by-task - ARC Challenge | 39.33 | Accuracy (%) | 33.5 |
| MMLU-by-task - Professional Medicine | 38.97 | Accuracy (%) | 32.2 |
| MMLU-by-task - Clinical Knowledge | 31.32 | Accuracy (%) | 32.1 |
Interactive version: theaggregate.ai/model?slug=pythia-12b-sft-v8-2-5k-steps · How It Works · Data refreshed daily, snapshot 2026-11-08.