3 ms·
Actually in this case they possibly are estimates. It's been known for some years[1] that LLMs do regression in-context. Frontier models have been trained agai
by nl 4mo ago
Actually in this case they possibly are estimates.
It's been known for some years[1] that LLMs do regression in-context. Frontier models have been trained against many, many issue text that include task break downs and estimates.
[1] https://arxiv.org/html/2409.04318v1 https://arxiv.org/html/2409.04318v1
- kube-system 4mo agoInteresting. So it may have learned how to estimate as a human but doesn’t understand that it doesn’t operate at that speed :D I wonder if there’s a reasonable way to give an llm parameters that give it a concept of its own execution speed. Seems that could be useful for multiple purposes
- nl 4mo agoYes, it's entirely possible to do that via RL. It'd be a fun little project you could do for less than $100 on a small LLM actually.