llama_ppo_1e6_new_tokenizerstep_8000: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1311 ± 35, rank #8529 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - GSM8K | 14.86 | Accuracy (%) (5-shot) | 36.7 |
| Open LLM Leaderboard v1 - HellaSwag | 78.64 | Normalized accuracy (%) (10-shot) | 35.1 |
| Open LLM Leaderboard v1 - WinoGrande | 74.03 | Accuracy (%) (5-shot) | 34.6 |
| Open LLM Leaderboard v1 - ARC Challenge | 54.78 | Normalized accuracy (%) (25-shot) | 32 |
| Open LLM Leaderboard v1 - MMLU | 46.63 | Accuracy (%) (5-shot) | 25.2 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 41.06 | MC2 (%) (0-shot) | 16 |
Interactive version: theaggregate.ai/model?slug=llama-ppo-1e6-new-tokenizerstep-8000 · How It Works · Data refreshed daily, snapshot 2026-11-08.