[megathread] LLM: Large Language Models
Prompt engineering | Instruction tuning | Chain of thought | Emergent properties
Unpredictable Abilities Emerging f. Large AI Models
https://www.quantamagazine.org/the-unpredictable-abilities-emerging-from-large-ai-models-20230316
Discussion: https://news.ycombinator.com/item?id=35195106
* LLM can display startling, unpredictable behaviors
* LLM model prompted to explain itself (a capacity called chain-of-thought reasoning) could correctly solve a math word problem; the same model without that prompt could not