Saul-Instruct-Clown-7B: benchmark results
An Arcee AI gradient SLERP merge of CorticalStack/pastiche-crown-clown-7b-dare-dpo and Equall/Saul-Instruct-v1, with separate attention and MLP interpolation schedules. Provider: Arcee AI. Access: Open.
Aggregate Rating 1442 ± 36, rank #4115 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - GSM8K | 67.85 | Accuracy (%) (5-shot) | 84.5 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 63.2 | MC2 (%) (0-shot) | 80.3 |
| Open LLM Leaderboard v1 - WinoGrande | 81.61 | Accuracy (%) (5-shot) | 80.1 |
| Open LLM Leaderboard v1 - HellaSwag | 86.23 | Normalized accuracy (%) (10-shot) | 79.7 |
| Open LLM Leaderboard v1 - ARC Challenge | 68.09 | Normalized accuracy (%) (25-shot) | 79.3 |
| Open LLM Leaderboard v1 - MMLU | 64.41 | Accuracy (%) (5-shot) | 69.8 |
Interactive version: theaggregate.ai/model?slug=saul-instruct-clown-7b · How It Works · Data refreshed daily, snapshot 2026-11-08.