falcon_1b_stage3: benchmark results
euclaise’s third-stage fine-tune, initialized from falcon_1b_stage2. The card leaves the training dataset unspecified; this release remains distinct from the earlier stages. Provider: euclaise. Access: Open.
Aggregate Rating 1140 ± 16, rank #10107 of 10916 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - Machine Learning | 35.71 | Accuracy (%) | 64.2 |
| MMLU-by-task - Econometrics | 28.07 | Accuracy (%) | 47.3 |
| MMLU-by-task - High School European History | 30.91 | Accuracy (%) | 34.3 |
| MMLU-by-task - Philosophy | 33.76 | Accuracy (%) | 34.2 |
| MMLU-by-task - Sociology | 30.85 | Accuracy (%) | 31.7 |
| MMLU-by-task - Logical Fallacies | 30.06 | Accuracy (%) | 31.4 |
| MMLU-by-task - US Foreign Policy | 34 | Accuracy (%) | 31.4 |
| MMLU-by-task - High School Microeconomics | 29.41 | Accuracy (%) | 30.3 |
| MMLU-by-task - Electrical Engineering | 28.97 | Accuracy (%) | 29.8 |
| MMLU-by-task - Anatomy | 31.85 | Accuracy (%) | 29.1 |
| MMLU-by-task - College Biology | 27.78 | Accuracy (%) | 28.9 |
| MMLU-by-task - Abstract Algebra | 26 | Accuracy (%) | 28.1 |
Interactive version: theaggregate.ai/model?slug=falcon-1b-stage3 · How It Works · Data refreshed daily, snapshot 2026-11-08.