Codesota · Models · Llama 2 70B (5-shot)6 results · 6 benchmarks
Model card
Llama 2 70B (5-shot).
unknown1 current SOTA
§ 02 · Benchmarks
Every benchmark Llama 2 70B (5-shot) has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | TriviaQA | Natural Language Processing · Question Answering | accuracy | 85.0% | #1 | — | source ↗ |
| 02 | Natural Questions | Natural Language Processing · Question Answering | accuracy | 33.0% | #2 | — | source ↗ |
| 03 | HumanEval+ | Computer Code · Code Generation | pass-1 | 29.9% | #9 | — | source ↗ |
| 04 | BIG-Bench Hard | Reasoning · Multi-step Reasoning | accuracy | 51.2% | #10 | — | source ↗ |
| 05 | GSM8K | Reasoning · Mathematical Reasoning | accuracy | 56.8% | #45 | — | source ↗ |
| 06 | MMLU | Reasoning · Commonsense Reasoning | accuracy | 68.9% | #55 | — | source ↗ |
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 03 · Strengths by area
Where Llama 2 70B (5-shot) actually performs.
§ 04 · Papers
1 paper with results for Llama 2 70B (5-shot).
- 2023-07-18· 6 results
Llama 2: Open Foundation and Fine-Tuned Chat Models
§ 06 · Sources & freshness
Where these numbers come from.
pwc-dump
6
results
0 of 6 rows marked verified.