Fetching the paper…

Logistic Regression makes small LLMs strong and explainable "tens-of-shot" classifiers · Around