The performance of large language models (LLMs) strongly depends on the temperature parameter. Empirically, at very low temperatures, LLMs generate sentences with clear repetitive structures, while at very high temperatures, generated sentences are often incomprehensible. In this study, using GPT-2, we numerically demonstrate that the difference between the two regimes is not just a smooth change but a phase transition with singular, divergent statistical quantities. Our extensive analysis shows that critical behaviors, such as a power-law decay of correlation in a text, emerge in the LLM at the transition temperature as well as in a natural language dataset. We also discuss that several statistical quantities characterizing the criticality should be useful to evaluate the performance of LLMs.