Llama 2 7B Chat Phr Mental Health 2048 [4bit]: benchmark results
Provider: Meta.
Aggregate Rating 1270 ± 37, rank #9064 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - ARC Challenge | 52.39 | Normalized accuracy (%) (25-shot) | 27 |
| Open LLM Leaderboard v1 - HellaSwag | 75.39 | Normalized accuracy (%) (10-shot) | 26.4 |
| Open LLM Leaderboard v1 - WinoGrande | 71.19 | Accuracy (%) (5-shot) | 26.1 |
| Open LLM Leaderboard v1 - GSM8K | 5.91 | Accuracy (%) (5-shot) | 24.9 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 42.89 | MC2 (%) (0-shot) | 21.9 |
| Open LLM Leaderboard v1 - MMLU | 39.77 | Accuracy (%) (5-shot) | 21 |
Interactive version: theaggregate.ai/model?slug=llama-2-7b-chat-phr-mental-health-2048-4bit · How It Works · Data refreshed daily, snapshot 2026-11-08.