Llama-3.1-8B-Instruct-abliterated_via_adapter: benchmark results
grimjim's Llama 3.1 8B Instruct with refusals removed by task-arithmetic-merging an abliteration LoRA extracted from failspy's Llama 3 8B abliterated-v3. Provider: Meta. Released 2024-07-25. Access: Open.
Unified ELO 1530 ± 1, rank #608 of 1553 rated models, from 7 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM Leaderboard | 44.67 | Average Score (%) | 89.7 |
| Open LLM Leaderboard - GPQA | 8.5 | Score | 70.7 |
| Open LLM Leaderboard - MATH Level 5 | 13.97 | Score | 60.2 |
| Open LLM Leaderboard - MMLU-Pro | 29.46 | Score | 57 |
| Open LLM Leaderboard - IFEval | 48.7 | Score | 56.9 |
| Open LLM Leaderboard - BBH | 29.42 | Score | 49.6 |
| Open LLM Leaderboard - MuSR | 9.26 | Score | 44.7 |
Interactive version: theaggregate.ai/model?slug=llama-3-1-8b-instruct-abliterated-via-adapter · How It Works · Data refreshed daily, snapshot 2026-09-08.