WestSeverus-10.7B: benchmark results
A 48-layer depth expansion by FelixChao of WestSeverus-7B-DPO-v2, concatenating layers 0-23 and 8-31. This is a modified checkpoint. Provider: FelixChao. Access: Open.
Aggregate Rating 1456 ± 39, rank #3503 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 72.3 | MC2 (%) (0-shot) | 93.3 |
| Open LLM Leaderboard v1 - ARC Challenge | 72.18 | Normalized accuracy (%) (25-shot) | 91.2 |
| Open LLM Leaderboard v1 - HellaSwag | 87.47 | Normalized accuracy (%) (10-shot) | 86.4 |
| Open LLM Leaderboard v1 - WinoGrande | 82.72 | Accuracy (%) (5-shot) | 85.2 |
| Open LLM Leaderboard v1 - MMLU | 65.06 | Accuracy (%) (5-shot) | 81.1 |
| Open LLM Leaderboard v1 - GSM8K | 62.17 | Accuracy (%) (5-shot) | 75.3 |
Interactive version: theaggregate.ai/model?slug=westseverus-10-7b · How It Works · Data refreshed daily, snapshot 2026-11-08.