e.star.7.b: benchmark results
e.star.7.b is a liminerity-published Mistral-family fine-tune of yam-peleg/Experiment26-7B. Its card credits gate369 and identifies Unsloth and Hugging Face TRL; publication attribution does not replace the card's development credit. Provider: liminerity. Access: Open.
Aggregate Rating 1422 ± 33, rank #5063 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - HellaSwag | 86.02 | Normalized accuracy (%) (10-shot) | 78.1 |
| Open LLM Leaderboard v1 - GSM8K | 61.18 | Accuracy (%) (5-shot) | 73.6 |
| Open LLM Leaderboard v1 - WinoGrande | 80.19 | Accuracy (%) (5-shot) | 72.8 |
| Open LLM Leaderboard v1 - ARC Challenge | 63.91 | Normalized accuracy (%) (25-shot) | 64.8 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 54.91 | MC2 (%) (0-shot) | 63.8 |
| Open LLM Leaderboard v1 - MMLU | 63.44 | Accuracy (%) (5-shot) | 60.7 |
Interactive version: theaggregate.ai/model?slug=e-star-7-b · How It Works · Data refreshed daily, snapshot 2026-11-08.