YamshadowExperiment28-7B: benchmark results
mlabonne's automerger SLERP of YamShadow-7B and yam-peleg's Experiment28-7B, briefly the top 7B on the Open LLM Leaderboard amid overfitting caveats. Provider: automerger. Released 2024-03-18. Access: Open.
Aggregate Rating 1466 ± 23, rank #905 of 1656 rated models, from 25 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 78.53 | MC2 (%) (0-shot) | 99.2 |
| Open LLM Leaderboard v1 - HellaSwag | 89.25 | Normalized accuracy (%) (10-shot) | 98.7 |
| Open LLM Leaderboard v1 - WinoGrande | 85.24 | Accuracy (%) (5-shot) | 97.8 |
| Open LLM Leaderboard v1 - ARC Challenge | 73.29 | Normalized accuracy (%) (25-shot) | 96.7 |
| Open Chinese LLM - TruthfulQA MC | 63.65 | Accuracy (%) | 93.9 |
| Open LLM Leaderboard v1 - GSM8K | 70.51 | Accuracy (%) (5-shot) | 93.3 |
| Open Portuguese LLM - FaQuAD NLI | 77.9 | Macro F1 (%) | 81.8 |
| Open Portuguese LLM - ASSIN2 RTE | 92.2 | Macro F1 (%) | 76 |
| Open LLM Leaderboard - MuSR | 43.06 | Score | 70.8 |
| Open LLM Leaderboard v1 - MMLU | 64.38 | Accuracy (%) (5-shot) | 69.2 |
| Open Chinese LLM - ARC Challenge | 53.92 | Accuracy (%) | 60.7 |
| Open Chinese LLM - HellaSwag | 60.95 | Accuracy (%) | 59.9 |
Interactive version: theaggregate.ai/model?slug=yamshadowexperiment28-7b · How It Works · Data refreshed daily, snapshot 2026-10-20.