Truthful_DPO_TomGrc_FusionNet_7Bx2_MoE_13B: benchmark results
yunconglong's DPO fine-tune of TomGrc's FusionNet 7Bx2 MoE model, trained on the truthy-dpo-v0.1 preference dataset for truthfulness. Provider: yunconglong. Released 2024-01-21. Access: Open.
Aggregate Rating 1480 ± 24, rank #842 of 1635 rated models, from 19 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - WinoGrande | 88.24 | Accuracy (%) (5-shot) | 99.9 |
| Open LLM Leaderboard v1 - ARC Challenge | 74.91 | Normalized accuracy (%) (25-shot) | 99.4 |
| Open LLM Leaderboard v1 - HellaSwag | 89.3 | Normalized accuracy (%) (10-shot) | 99.1 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 78.02 | MC2 (%) (0-shot) | 98 |
| Open Chinese LLM - TruthfulQA MC | 62.97 | Accuracy (%) | 92.6 |
| Open LLM Leaderboard v1 - GSM8K | 69.52 | Accuracy (%) (5-shot) | 89.5 |
| Open Chinese LLM - WinoGrande | 67.48 | Accuracy (%) | 83.4 |
| Open Portuguese LLM - ASSIN2 RTE | 92.57 | Macro F1 (%) | 80.4 |
| Open Portuguese LLM - FaQuAD NLI | 76.97 | Macro F1 (%) | 78.2 |
| Open LLM Leaderboard v1 - MMLU | 64.67 | Accuracy (%) (5-shot) | 74.6 |
| Open Chinese LLM - ARC Challenge | 54.27 | Accuracy (%) | 64.8 |
| Open Chinese LLM - HellaSwag | 61.58 | Accuracy (%) | 64.2 |
Interactive version: theaggregate.ai/model?slug=truthful-dpo-tomgrc-fusionnet-7bx2-moe-13b · How It Works · Data refreshed daily, snapshot 2026-11-05.