I have a project with an LLM (small model) and need help verifying prompts. https://jeremyberman.substack.com/p/how-i-got-a-record-536-on-arc-agi https://jeremyberman.substack.com/p/how-i-got-the-highest-score-on-arc-agi-again It is the same as the links, but we are using small language models and going through the prompts to understand when the models fail to reason. We can adjust the prompts to improve accuracy. So, the importance of understanding why smaller models fail, how to improve them, and the changes in the prompts and the evaluation process. - Writing the process in a report