falcon_1b_stage2: benchmark results
euclaise’s second-stage fine-tune, initialized from falcon_1b_stage1. The public card does not identify the training dataset; the stage is retained as a separate checkpoint. Provider: euclaise. Access: Open.
Aggregate Rating 1128 ± 18, rank #10362 of 10916 rated models, from 62 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMLU-by-task - College Physics | 28.43 | Accuracy (%) | 67.4 |
| MMLU-by-task - Global Facts | 32 | Accuracy (%) | 44.3 |
| MMLU-by-task - Moral Scenarios | 24.36 | Accuracy (%) | 32.6 |
| MMLU-by-task - Business Ethics | 33 | Accuracy (%) | 32 |
| MMLU-by-task - Machine Learning | 28.57 | Accuracy (%) | 30.8 |
| MMLU-by-task - High School Mathematics | 25.56 | Accuracy (%) | 29.5 |
| MMLU-by-task - College Biology | 27.78 | Accuracy (%) | 28.9 |
| MMLU-by-task - TruthfulQA MC1 | 24.6 | Accuracy (%) | 28.9 |
| MMLU-by-task - Marketing | 29.49 | Accuracy (%) | 28 |
| MMLU-by-task - HellaSwag | 48.87 | Accuracy (%) | 27.5 |
| MMLU-by-task - Computer Security | 31 | Accuracy (%) | 27.3 |
| MMLU-by-task - Conceptual Physics | 31.06 | Accuracy (%) | 27.3 |
Interactive version: theaggregate.ai/model?slug=falcon-1b-stage2 · How It Works · Data refreshed daily, snapshot 2026-11-08.