LLaMA-65B
LLaMA-65B places in the top three on HELM Classic.
Meta Released 24 Feb 2023 13
Benchmarks with a score
0
Ranked first, out of five or more models
1
In the top three
Meta Feb 2023 LLaMA-65B
Open weights
110.1
#225 of 271
Epoch Capabilities Index
- LAMBADA 77.7% #7 /26
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| TriviaQA | 86% | #5 of 38 | Epoch AI | |
| HellaSwag | 84.2% | #12 of 74 | Epoch AI | |
| BoolQ | 87.1% | #16 of 75 | Epoch AI | |
| LAMBADA | 77.7% | #7 of 26 | Epoch AI | |
| PIQA | 82.8% | #16 of 58 | Epoch AI | |
| ARC-Challenge | 69.5% | #23 of 75 | Epoch AI | |
| OpenBookQA | 60.2% | #14 of 41 | Epoch AI | |
| WinoGrande | 77% | #28 of 76 | Epoch AI | |
| BIG-Bench Hard | 58.4% | #17 of 45 | Epoch AI | |
| MMLU | 63.4% | #76 of 127 | Epoch AI |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GSM8K | 54.4% | #33 of 82 | Epoch AI |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| HELM Classic | 90.8 | #2 of 17 | Official leaderboard | |
| HELM Lite | 34.5 | #54 of 71 | Official leaderboard |
Meta's models over time
Meta's models in color, against every other model on the Epoch Capabilities Index, by release date.