3 ms·
>Chain of reasoning is a lot of context to guide token generation, but we simply see that newer models don’t need that context to get to the answer I thought e
by irthomasthomas 3mo ago
>Chain of reasoning is a lot of context to guide token generation, but we simply see that newer models don’t need that context to get to the answer
I thought each new generation typically used more reasoning tokens?
- general_reveal 3mo agoThey do if they are a reasoning-variant. That doesn’t necessarily mean it actually needed to reason for many questions, your prompt + regular context could be enough to get a good answer compared to prior models where you’d absolutely have to put it into a reasoning-loop to get an accurate answer. It’s on by default, in a way. You can probably prompt these models with “and don’t reason about it, just give me the answer” and probably get a comparably good response without it using reasoning tokens for many things.