List your company

Email Writing on RefuteBench Email Writing 1.0 (test)

98FAGPT-4
63.6872.5981.590.41Feb 21, 2024

Evaluation results

MethodLinks
GPT-42024.02988159704272.9
Claude-22024.02956839541736.2
ChatGPT2024.02924730501129.3
LLAMA-2-7B-ChatParameters=7B, Model Type=Chat2024.02841224112.25
LLAMA-2-13B-ChatParameters=13B, Model Type=Chat2024.028242222217.25
Mistral-7B-Instruct-v0.2Parameters=7B, Model Type=Instruct2024.02765031342035.35
ALPACA-7BParameters=7B2024.026525111013.1