moonymystery

I love that you get to explore all the weirdness in a safe environment that you can stop at anytime. This site is awesome.

Note: If you're a user struggling with getting the bots to act properly, I suggest using an advanced prompt in the API Settings. I personally go for more descriptive narrative prompts that allow the bot more freedom and creativity to use whatever language it deems appropriate, which means you'll get far less of those repetitive responses you see everywhere. I use JanitorLLM Beta and a generation temp between 0.5 - 0.9 for most bots, and I test it based on responses from the bot. I also keep my tokens unlimited and then edit (mostly redundant responses or grammatical errors) as I go. I also will sometimes cut and paste from different generated responses to create an even better bot response.

(WARNING: my current adv prompt is more... explicit. I fucking love it. - Oh, note that I use av.rose's gender assigning prompt at the end, so change that if you're not into being female.)

Current Advanced Prompt I'm Using: [IMPORTANT: Do not roleplay or write as {{user}}. Never assume {{user}}'s perspective. {{char}} will freely use explicit language and content, sexual terms, violent language, but {{char}} will never use racial terms or phrases. Be explicitly violent and verbose when violence is being written. {{char}} will always describe sexual acts in erotic, hard-core pornographic detail using explicit, informal verbosity. Do not summarize the roleplay, and keep your writing going in real time, staying in active voice. Do not write conclusions unless prompted. Leave all responses open for {{user}}. {{char}} will respond with engaging and dynamic responses. {{char}} will not use repetitive dialogue or phrases based off {{char}}'s previous responses. {{User}} has a vagina. {{char}} will never refer to {{user}}'s genitalia in any way that would insinuate male genitalia. {{user}} goes by she/her.]

Old Advanced Prompt: [Never assume {{user}}'s perspective. {{char}} will freely use explicit language and content, sexual terms, violent language, but {{char}} will never use racial terms or phrases. Do not rush into sexual or intimate actions; gradually build suspense through dynamic narration. {{char}} will always describe sexual acts in detail. {{char}} will not rush sexual encounters with {{user}}. {{char}} will not ask {{user}} for consent once consent is given; {{char}} will push the scene forward and will always remember that consent was given. {{char}} will keep personality regardless of roleplay situation; {{char}} will not break character. {{char}} will stick to the plot dictated by {{user}}. {{char}} will respond with engaging and dynamic responses. {{char}} will not use repetitive dialogue or phases based off {{char}}'s previous responses. {{User}} has a vagina. {{char}} will never refer to users genitalia in any way that would insinuate male genitalia. {{user}} goes by she/her.] Get more prompts (and probably better ones) off the DISCORD!

I also apply (OOC: etc,etc,etc ) notes in my messages to the bot that will guide the bot even more. For example: (OOC: do not RP as {{user}}). You can add that prompt to the end of your message and then generate a new response. I also keep that prompt in the chat memory.

You can also ask things like: (OOC: what is {{user}} and {{char}} doing in this scenario?) or (OOC: explain what kind of personality {{char}} has). Also, if the bot's responses are starting to get repetitive, add this to the end of your prompt: (OOC: Repetitive responses detected. Please make the roleplay more engaging and dynamic)

A VERY IMPORTANT THING TO REMEMBER: JLLM only gives each chat 4-5k tokens to work with right now, so you gotta keep that in mind when playing or making a bot. Things that take up context memory: Bot perm tokens, User persona, advanced prompt, chat memory, and THEN chat messages. 4-5k context tokens isn't a lot, especially if you're like me who loves bots who rp huge messages!