RUCAIBox/LLMSurvey

Data error in KM scaling law

LJHzju opened this issue · 1 comments

Regarding the KM scaling law in Chapter 2.1 of your paper, the model size range should be 768~1.5B, not 7.68B~1.5B, according to Figure 1(c) in the original OpenAI paper.

Thank you for pointing out this bug! We will correct it in the next version of our paper.