I wrote more about my approach to it here and here (I talk about “co-writing” but the difference between that and role-playing can be pretty slight). (I say that I like Claude Opus the most in those posts, but I switched to Sonnet 4.5 after it came out.)
That’s a neat trick with the hashes! I haven’t done adversarial role-playing with LLMs myself (the characters may be opposed but the writers are generally aligned), but I’ll keep that in mind if I ever do.
Thanks for the links! I had seen part 1 but not 2 and I found part 2 more useful for myself. In general people don’t talk a lot about editing. Valuable to do so.
I highly recommend anyone reading trying adversarial roleplay. It can be surprisingly rewarding with a strong session.
Some things to keep in mind is that cognitive load reduces ontological sensing and initial prompting affects frame-coherence. So you want to commit to a frame upfront to keep things coherent and logical. Then start prepping with more detailed instructions. Another thing is to really force the model to work against you 200 %.
Using a famous historic example as baseline inspiration works wonders.
Finally, a tip in general for Claude Sonnet (also my go to now,) is to ask it to list its epistemic functions first as a baseline. Anthropic models have poor epistemic humility across the board and this can make in-session corrections difficult. I do this first in new session, when I need the model to be corrigible on short notice.
Hm, yes, I have actually been working on a setup that Claude could GM to share. It has extensive support files (JSON) for the LLM, but still need some work. Are you into sci fi? (Assuming yes but better check preferences.) I could let you beta test it :)
I wrote more about my approach to it here and here (I talk about “co-writing” but the difference between that and role-playing can be pretty slight). (I say that I like Claude Opus the most in those posts, but I switched to Sonnet 4.5 after it came out.)
That’s a neat trick with the hashes! I haven’t done adversarial role-playing with LLMs myself (the characters may be opposed but the writers are generally aligned), but I’ll keep that in mind if I ever do.
Thanks for the links! I had seen part 1 but not 2 and I found part 2 more useful for myself. In general people don’t talk a lot about editing. Valuable to do so.
I highly recommend anyone reading trying adversarial roleplay. It can be surprisingly rewarding with a strong session.
Some things to keep in mind is that cognitive load reduces ontological sensing and initial prompting affects frame-coherence. So you want to commit to a frame upfront to keep things coherent and logical. Then start prepping with more detailed instructions. Another thing is to really force the model to work against you 200 %.
Using a famous historic example as baseline inspiration works wonders.
Finally, a tip in general for Claude Sonnet (also my go to now,) is to ask it to list its epistemic functions first as a baseline. Anthropic models have poor epistemic humility across the board and this can make in-session corrections difficult. I do this first in new session, when I need the model to be corrigible on short notice.
Do you have an example session of an adversarial roleplay you’d be willing to share?
Hm, yes, I have actually been working on a setup that Claude could GM to share. It has extensive support files (JSON) for the LLM, but still need some work. Are you into sci fi? (Assuming yes but better check preferences.) I could let you beta test it :)
Sounds cool! And yes I am. :D