Remix.run Logo
ekianjo 4 hours ago

prompts will give very different results. this is where you do the work.

cryo32 4 hours ago | parent | next [-]

I disagree. The LLM outputs really do lack anything original or interesting. They just produce banal copy whatever you ask them.

A good editor could probably reduce all LLM outputs on a subject down to the same point.

Mikhail_Edoshin 2 hours ago | parent | prev | next [-]

A controller has to be at least as complex as what it is supposed to control.

roncesvalles 4 hours ago | parent | prev | next [-]

Yes but not very different results (unless you're adding new information to your prompt or reducing some ambiguity). Prompt engineering is mostly pseudoscience.

zarzavat 4 hours ago | parent [-]

What we need is steering so that we can have models with different personalities, not just different prompts (because context is subject to forgetting), but this will never happen with closed-weight models, I'm not sure if it's even feasible at scale.

Yet another reason why the future is open weight.

hansmayer 3 hours ago | parent | prev [-]

[dead]