Llama 2 7B Chat Activity Fine Tuned V4 [bfloat16] [rev: C964836b5748]: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1319 ± 33, rank #8340 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - GSM8K | 19.26 | Accuracy (%) (5-shot) | 40.5 |
| Open LLM Leaderboard v1 - WinoGrande | 73.95 | Accuracy (%) (5-shot) | 34.2 |
| Open LLM Leaderboard v1 - HellaSwag | 78.1 | Normalized accuracy (%) (10-shot) | 32.9 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 45.77 | MC2 (%) (0-shot) | 32.2 |
| Open LLM Leaderboard v1 - ARC Challenge | 54.27 | Normalized accuracy (%) (25-shot) | 31 |
| Open LLM Leaderboard v1 - MMLU | 48.44 | Accuracy (%) (5-shot) | 27.7 |
Interactive version: theaggregate.ai/model?slug=llama-2-7b-chat-activity-fine-tuned-v4-bfloat16-rev-c964836b5748 · How It Works · Data refreshed daily, snapshot 2026-11-08.