• OpenAI API Key

• Beginners
• Intermediate users familiar with basic general worksheets concepts

This file presents a comprehensive evaluation of GPT-3.5 Turbo's reasoning abilities using a dataset of academic prompts published in the latter half of 2023.

The evaluation focuses on analyzing the model's responses to various reasoning tasks using different prompting techniques, such as zero-shot-CoT, Prompt and plan, Analogical Prompting, and Emotion Prompt.

The results provide insights into the model's strengths and weaknesses in reasoning.

In the Tester sheet, cell A2, insert your API KEY. From the Evaluator menu, select Generate Test.