Experiment23-7B: benchmark results
A 32-layer Mistral-family checkpoint released as Experiment23-7B. 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 1461 ± 43, rank #3278 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 78.87 | MC2 (%) (0-shot) | 99.4 |
| Open LLM Leaderboard v1 - WinoGrande | 85.32 | Accuracy (%) (5-shot) | 98.1 |
| Open LLM Leaderboard v1 - HellaSwag | 88.77 | Normalized accuracy (%) (10-shot) | 94 |
| Open LLM Leaderboard v1 - ARC Challenge | 72.35 | Normalized accuracy (%) (25-shot) | 91.7 |
| Open LLM Leaderboard v1 - GSM8K | 62.4 | Accuracy (%) (5-shot) | 75.7 |
| Open LLM Leaderboard v1 - MMLU | 64.17 | Accuracy (%) (5-shot) | 66.4 |
Interactive version: theaggregate.ai/model?slug=experiment23-7b · How It Works · Data refreshed daily, snapshot 2026-11-08.