TinyStories: How Small Can Language Models Be and Still Speak Coherent English? (2023)

Re:TinyStories: How Small Can Language Models Be and Still S... by SketchySeaBeast on Thursday January 02, 2025 @03:09PM (#42578282)

But what if they want to know they are finding the correct answer?

[ Parent | on HN ]

Re:TinyStories: How Small Can Language Models Be and Still S... by elliotto on Thursday January 02, 2025 @04:16PM (#42579008)

Asking anonymous people on a forum would be much better.

[ Parent | on HN ]

Re:TinyStories: How Small Can Language Models Be and Still S... by SketchySeaBeast on Thursday January 02, 2025 @05:00PM (#42579404)

At least a forum with domain-specific knowledge.

[ Parent | on HN ]

Re:TinyStories: How Small Can Language Models Be and Still S... by gpm on Thursday January 02, 2025 @05:07PM (#42579466)

And people to go "no, that's wrong" if someone posts something that's wrong.

[ Parent | on HN ]

Re:TinyStories: How Small Can Language Models Be and Still S... by greesil on Thursday January 02, 2025 @07:49PM (#42580867)

No, any answer will do.

[ Parent | on HN ]

Re:TinyStories: How Small Can Language Models Be and Still S... by cauliflower2718 on Friday January 03, 2025 @06:55AM (#42584850)

I think basic definitions for LLMs are solidly within the bounds of what we would expect e.g. chatgpt to be competent at. The task (defining terms) is simple and the specific content (basic LLM stuff) is easy to check by anyone who works on the LLM.

I agree with the general sentiment that we should not just blindly trust LLMs though.

[ Parent | on HN ]