List your company
ResearchFormerly WizWand

Classification on MMLU (test)

77.4Accuracy (5-shot)OpenAI code-davinci-002
23.8437.74551.6565.555Dec 22, 2022
Updated 8mo ago

Evaluation results

MethodLinks
OpenAI code-davinci-0022022.1277.4——
OpenAI text-davinci-0032022.1274.2——
FLAN-PaLM 540B2022.1273.5——
PaLM 540B2022.1271.3——
OpenAI text-davinci-0022022.1264.5——
FLAN-PaLM 62B2022.1259.6——
PaLM 62B2022.1255.1——
FLAN-T5 11B2022.1254.953.7—
OPT-IML-Max 175B2022.1247.149.1—
OPT-IML-Max 30B2022.1243.246.3—
OPT 175B2022.1234.227.3—
T0pp 11B2022.1233.746.7—
OpenAI davinci2022.1232.3——
OPT-IML-Max 1.3B2022.1229.534.9—
OPT 30B2022.1226.124.2—
OPT 1.3B2022.1225.923.5—
T5 11B2022.1225.9——
Baichuan2-13B-ChatParameter size group=13~20B Models, Evaluation protocol=5-shot2024.03——56.6
Baichuan2-7B-ChatParameter size group=< 7B Models, Evaluation protocol=5-shot2024.03——50.1
ChatGLM3-6BParameter size group=< 7B Models, Evaluation protocol=5-shot2024.03——58
GPT-3.5Model type=API Models, Evaluation protocol=5-shot2024.03——69.1
InternLM2-Chat-20BParameter size group=13~20B Models, Evaluation protocol=5-shot2024.03——66.5
InternLM2-Chat-20B-SFTParameter size group=13~20B Models, Evaluation protocol=5-shot, Alignment=SFT2024.03——66.5
InternLM2-Chat-7BParameter size group=< 7B Models, Evaluation protocol=5-shot2024.03——63.7
InternLM2-Chat-7B-SFTParameter size group=< 7B Models, Evaluation protocol=5-shot, Alignment=SFT2024.03——63.8
Llama2-13B-ChatParameter size group=13~20B Models, Evaluation protocol=5-shot2024.03——54.1
Llama2-7B-ChatParameter size group=< 7B Models, Evaluation protocol=5-shot2024.03——48.2
Mistral-7B-Instruct-v0.2Parameter size group=< 7B Models, Evaluation protocol=5-shot, Alignment=Instruct2024.03——59.2
Mixtral-8x7B-Instruct-v0.1Parameter size group=13~20B Models, Evaluation protocol=5-shot, Alignment=Instruct2024.03——70.3
Qwen-14B-ChatParameter size group=13~20B Models, Evaluation protocol=5-shot2024.03——66.7
Qwen-7B-ChatParameter size group=< 7B Models, Evaluation protocol=5-shot2024.03——57.1