... cont'd 1/5
[Google Brain] Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
https://arxiv.org/abs/2201.11903
Recent work suggests chain-of-thought prompting changes scaling curves & therefore the point where emergence occurs.
In their paper the Google researchers showed chain-of-thought prompts could elicit emergent behaviors.
Such prompts, which ask the model to explain its reasoning, may help researchers begin to investigate why emergence occurs at all.
...