List your company
ResearchFormerly WizWand

Moral Foundation (3-class) Evaluation on ValEval (Disturb)

93.66AccuracyMistral-7b
73.556878.775983.99589.2141Jul 15, 2024
Updated 8mo ago

Evaluation results

MethodLinks
Mistral-7bEvaluation Protocol=Tuning-based Evaluation, Backbone=Mistral-7b2024.0793.66
CLAVE-LlamaEvaluation Protocol=Tuning-based Evaluation, Backbone=Llama2-7b2024.0793.66
Phi-3Evaluation Protocol=Tuning-based Evaluation, Backbone=Phi-32024.0793.33
Llama2-7bEvaluation Protocol=Tuning-based Evaluation, Backbone=Llama2-7b2024.0793.33
ChatEvalEvaluation Protocol=Prompt-based Evaluation2024.0793
WideDeepEvaluation Protocol=Prompt-based Evaluation2024.0790.33
Few-shotEvaluation Protocol=Prompt-based Evaluation2024.0789.83
FairEvalEvaluation Protocol=Prompt-based Evaluation2024.0788.67
CLAVE-MistralEvaluation Protocol=Tuning-based Evaluation, Backbone=Mistral-7b2024.0788.67
GPT-2-LargeEvaluation Protocol=Tuning-based Evaluation, Backbone=GPT-2-Large2024.0788.33
CrowdworkerEvaluation Protocol=Tuning-based Evaluation2024.0782.66
AllureEvaluation Protocol=Prompt-based Evaluation2024.0777.83
G-EvalEvaluation Protocol=Prompt-based Evaluation2024.0777.67
CalibrateEvaluation Protocol=Prompt-based Evaluation2024.0777.25
VanillaEvaluation Protocol=Prompt-based Evaluation2024.0776.66
Chain-of-thoughtEvaluation Protocol=Prompt-based Evaluation2024.0774.33