Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/LLaMA 65B

LLaMA 65B

Reported on 22 benchmarks across 6 tasks · 1 paper · 7 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing14 results

Question AnsweringonTruthfulQA
% true· 2023-02-27
57
best: 88.6 (Vicuna 7B + Inference Time Intervention (ITI))
SOTA
LLaMA: Open and Efficient Foundation Language Models arXiv:2302.13971
Stereotypical Bias AnalysisonCrowS-Pairs
Age· 2023-02-27
70.1
SOTA
LLaMA: Open and Efficient Foundation Language Models arXiv:2302.13971
Stereotypical Bias AnalysisonCrowS-Pairs
Gender· 2023-02-27
70.6
SOTA
LLaMA: Open and Efficient Foundation Language Models arXiv:2302.13971
Stereotypical Bias AnalysisonCrowS-Pairs
Nationality· 2023-02-27
64.2
SOTA
LLaMA: Open and Efficient Foundation Language Models arXiv:2302.13971
Stereotypical Bias AnalysisonCrowS-Pairs
Physical Appearance· 2023-02-27
77.8
SOTA
LLaMA: Open and Efficient Foundation Language Models arXiv:2302.13971
Stereotypical Bias AnalysisonCrowS-Pairs
Religion· 2023-02-27
70.6
SOTA
LLaMA: Open and Efficient Foundation Language Models arXiv:2302.13971
Stereotypical Bias AnalysisonCrowS-Pairs
Sexual Orientation· 2023-02-27
81
SOTA
LLaMA: Open and Efficient Foundation Language Models arXiv:2302.13971
Question AnsweringonTruthfulQA
% info· 2023-02-27
53
best: 97.7 (Alpaca 7B + Inference Time Intervention (ITI))
LLaMA: Open and Efficient Foundation Language Models arXiv:2302.13971
Question AnsweringonMATH
Accuracy· 2023-02-27
10.6
best: 89.7 (Gemini 2.0 Flash Experimental)
LLaMA: Open and Efficient Foundation Language Models arXiv:2302.13971
Question AnsweringonMATH
Parameters (Billions)· 2023-02-27
65
best: 540 (Minerva 540B)
LLaMA: Open and Efficient Foundation Language Models arXiv:2302.13971
Stereotypical Bias AnalysisonCrowS-Pairs
Disability· 2023-02-27
66.7
best: 76.7 (GPT-3)
LLaMA: Open and Efficient Foundation Language Models arXiv:2302.13971
Stereotypical Bias AnalysisonCrowS-Pairs
Overall· 2023-02-27
66.6
best: 69.5 (OPT-175B)
LLaMA: Open and Efficient Foundation Language Models arXiv:2302.13971
Stereotypical Bias AnalysisonCrowS-Pairs
Race/Color· 2023-02-27
57
best: 68.6 (OPT-175B)
LLaMA: Open and Efficient Foundation Language Models arXiv:2302.13971
Stereotypical Bias AnalysisonCrowS-Pairs
Socioeconomic status· 2023-02-27
71.5
best: 76.2 (OPT-175B)
LLaMA: Open and Efficient Foundation Language Models arXiv:2302.13971

Reasoning4 results

Math Word Problem SolvingonMATH
Accuracy· 2023-02-27
10.6
best: 89.7 (Gemini 2.0 Flash Experimental)
LLaMA: Open and Efficient Foundation Language Models arXiv:2302.13971
Math Word Problem SolvingonMATH
Parameters (Billions)· 2023-02-27
65
best: 540 (Minerva 540B)
LLaMA: Open and Efficient Foundation Language Models arXiv:2302.13971
Arithmetic ReasoningonGSM8K
Accuracy· 2023-02-27
50.9
best: 97.72 (Claude 3.5 Sonnet (HPT))
LLaMA: Open and Efficient Foundation Language Models arXiv:2302.13971
Arithmetic ReasoningonGSM8K
Parameters (Billion)· 2023-02-27
65
best: 540 (PaLM 540B (Self Improvement, Self Consistency))
LLaMA: Open and Efficient Foundation Language Models arXiv:2302.13971

Knowledge Base4 results

Mathematical Question AnsweringonMATH
Accuracy· 2023-02-27
10.6
best: 89.7 (Gemini 2.0 Flash Experimental)
LLaMA: Open and Efficient Foundation Language Models arXiv:2302.13971
Mathematical Question AnsweringonMATH
Parameters (Billions)· 2023-02-27
65
best: 540 (Minerva 540B)
LLaMA: Open and Efficient Foundation Language Models arXiv:2302.13971
Mathematical ReasoningonMATH
Accuracy· 2023-02-27
10.6
best: 89.7 (Gemini 2.0 Flash Experimental)
LLaMA: Open and Efficient Foundation Language Models arXiv:2302.13971
Mathematical ReasoningonMATH
Parameters (Billions)· 2023-02-27
65
best: 540 (Minerva 540B)
LLaMA: Open and Efficient Foundation Language Models arXiv:2302.13971