Experiment2-7B: benchmark results
A 40-layer Mistral-family checkpoint released as Experiment2-7B. The exact evaluated configuration uses 32,000 vocabulary entries; the release does not document a reproducible training recipe or establish equivalence with other numbered experiments. Provider: Other. Access: Open.
Aggregate Rating 1454 ± 38, rank #3591 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - ARC Challenge | 72.18 | Normalized accuracy (%) (25-shot) | 91.2 |
| Open LLM Leaderboard v1 - HellaSwag | 88.15 | Normalized accuracy (%) (10-shot) | 89.8 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 69.97 | MC2 (%) (0-shot) | 89.8 |
| Open LLM Leaderboard v1 - MMLU | 65.1 | Accuracy (%) (5-shot) | 81.7 |
| Open LLM Leaderboard v1 - WinoGrande | 81.22 | Accuracy (%) (5-shot) | 77.8 |
| Open LLM Leaderboard v1 - GSM8K | 63.68 | Accuracy (%) (5-shot) | 77.3 |
Interactive version: theaggregate.ai/model?slug=experiment2-7b · How It Works · Data refreshed daily, snapshot 2026-11-08.