A research team has just finished the pre-training phase for a new large language model, using a massive corpus of text and code from the public internet. Before beginning any instruction fine-tuning, a researcher tests the model with the following prompt:

`Text: 'The sun is a star at the center of the Solar System. It is a nearly perfect sphere of hot plasma, with internal convective motion that generates a magnetic field via a dynamo process.'

Summarize the preceding text in one sentence.`

To their surprise, the model responds with: 'The sun is a plasma star at the center of our solar system that generates a magnetic field.'

Based on the principles of how models learn during pre-training, provide the most likely explanation for why the model was able to perform this summarization task without any explicit fine-tuning.