WORLDSIM (Alpha) for LATTE
MINOR UPDATE: OCT 14
pancatstack (Gourmet) for Claude2 pancatstack (CIMB) for Claud3 Credits Instructions Download Modifications How to use the Companion Lorebook Known Issues Troubleshooting Contact/Support Me
WORLDSIM is good for handling a wide variety of characters all the way from wholesome to dead dove and keeping them distinct, in character, and realistic.
Optimized with lorebooks in mind and side character friendly + implements them IMMERSIVELY.
THIS PRESET IS INTENDED FOR USE WITH chatgpt-4o-latest ONLY. I have no clue if it works with others.
CREDITS
I studied a bunch of different prompts and other creators' notes about JBs and CoT so I couldn't possibly attribute everyone but the main ones are:
Avani Currently using an adapted version of her JB method, hehe. Her JB is also highly recommended by most people that I've talked to so yea go check her's out!
KaruKaru Her JB method helped me develop my preset! I might have given up if not for that because LATTE's filter is ridiculous and I was frankly spoiled by how easy it was to JB Claude via Prefill 🙃 So yea definitely go check her stuff out because her brain is humongous. You can find the scyllacord thread for her LATTE JB here, at this time she does not have a rentry for it.
rarestMeow rentry 1 and rentry 2 learned frankly most that I know about CoT and LLMs thanks to his documentation. Highly recommended to check out his rentry if you want to learn how to best figure this shit out yourself.
Anonymous creator of the logit bias they did not want to be credited, but the logit bias was 99% their work. I just added the refusals and the < so ye
ALSO CHECK OUT
CharProvider I use his format for all my bots and am enjoying it the most so far. HIGHLY RECOMMENDED to convert your favorite bots to this format. If you have an assistant or character creation bot, you can have them convert the card for you.
Mega Converter Extremely convenient JB you can use to convert cards to JED or create new ones!
INSTRUCTIONS
1) download attached JSON.
2) if your ST isn't 1.12.6 go update it you silly.
3) IMPORT REQUIRED. Do NOT try copy-pasting the settings by yourself WORLDSIM has a VERY. SPECIFIC. ORDER.
4) Scroll down and IMPORT the included Logit Bias file
USE OF THE INCLUDED LOGIT BIAS IS REQUIRED FOR PROPER FUNCTIONALITY
5) Go into Regex Settings and Import ALL of the included regex files
USE OF ALL INCLUDED REGEXES IS REQUIRED FOR PROPER FUNCTIONALITY
6) Go into World Info and Import the included companion lorebook
USE OF THE INCLUDED LOREBOOK IS REQUIRED FOR PROPER FUNCTIONALITY
DOWNLOAD
.zip file includes:
Folder including the preset .json
Folder including the REQUIRED logit bias .json
Folder including ALL REQUIRED regex script .json
Folder including the REQUIRED companion lorebook .json
Update Changes: Now using Avani's JB method since it seems more consistent Re-added 💭Thinking Minor structural changes Big changes to ↓Char functionality in the lorebook, more ways to modify the preset. Some of which are Avani's prompts!
MODIFICATIONS
How the fuck do I use this stupid ass lorebook
If you're lazy, all you have to do is set it as a Global Lorebook for all chats and call it a day, if you use Character {{char}} cards.
If you are using a SETTING {{char}} card, however, you will have to change the toggles to this:
Modification is pretty straightforward, just make a duplicate of the lorebook and customize it to whatever chat you want, following the guidelines I've made.
If you have a bajillion characters in your one specific chat I'm sorry if I didn't give you enough space, you will have to bump up the Order range to fit them all in whatever category they belong. Anyways, the most important things you'll want to take a look at are highlighted in these boxes:
First you will want to change the sorting to Custom so that it looks readable to your eyeballs. Then start shoving stuff in there. 🟢 Entries are keyword-triggered, meaning the entry won't show in the prompt unless the defined keyword(s) appears at X depth in the Chat History. 🔵 Entries are constant, and will always be sent to the prompt as long as they are active. I don't generally recommend use of this unless it's like for a main character or the setting that you are currently in. The Order column signifies what their placement will be when sent to the prompt, so if John has an order of 1, Kelly has an order of 2, and Bobby has an order of 3, assuming they are all active the prompt will read them in this order: John, Kelly, Bobby. If you split up characters into a few separate entries (for like, NSFW defs or Example Dialog) then you will want to keep a close eye on this so you don't accidentally set John's cock description in the middle of Kelly's tags, for example.
And btw change your Activation settings to this:
Or at least the Context % and Budget Cap, because if they are too low then the lorebook entries won't trigger. If you notice that certain entries aren't being sent despite being toggled on and SHOULD be triggering, this is the first place you should check! (Budget Cap of 0 means no set limit btw and it will ONLY be based off of your Context %)
• I wanna move prompts around—general rule, DO NOT MOVE ANYTHING. If you must, OPEN THE INDIVIDUAL PROMPTS SO YOU ARE COGNIZANT OF WHICH XML TAGS I USE AND WHERE THEY ARE LOCATED, all XML tags must be wrapped appropriately or you risk breaking the preset.
• I want shorter/longer messages—Word Count is found under CHAT MODIFIERS in the lorebook, if it doesn't work lmk.
• I want to use a CoT but don't really like this one/certain parts aren't relevant to my RP/it uses too many tokens—If you want to use a different one, just make a custom prompt, drag it next to my Thinking, copy-paste it, and remove the contents of my <thinking> and add your own or one from whatever other prompt you like. If you use it and it doesn't order the output correctly, toggle on 📝Reminder and edit it so that the output will be ordered in the way that you like.
• I don't like Info Board how do I modify or get rid of it—Basically same as above, open the prompt to modify as you see fit. Or if you don't want it, toggle it off entirely. If you have 📝Reminder toggled on then remove reference to the Info Board inside of it.
KNOWN ISSUES
Message is blank sometimes!
Means you hit the filter. Turn on the gaslight prompts. You will still get the occasional blank, that's just how LATTE is in the current meta of JBing. If you get LOTS of blanks, however, make sure you have the Logit Bias enabled. Another thing you can try is turning Streaming (found above the Temperature slider) on or off, whatever the opposite setting of what you have it on is. Normally I have it turned on and it works for me for approx 60-70% of swipes of extreme NSFW content but I've had people come to me saying toggling Streaming off helps them a lot. Last thing I suggest trying is turning 💭Thinking off if it isn't already, for me that produces more consistent successful gens.
Sometimes doesn't produce <thinking>
Open 📝Reminder and edit it to include 💭Thinking in the position that you want it, then toggle it on. Try swiping now.
Looping & Repetition
Seems to be exacerbated with 💭Thinking, if you experience this yourself just try switiching it on and off from time to time. Though frankly you don't really need to use it unless the LLM is struggling with logic.
TROUBLESHOOTING
NOTE: If you moved ANYTHING in my preset without knowing what you're doing and it's fucking up do not complain to me 😸👍
I'm getting blank responses!
Toggle on ONE of the Anti-Refusals and keep trying.
The bot is speaking for me!
This can happen for multiple reasons:
- The character card (description, Scenario, and/or intro message). If any of those act for, speak for, or even IMPLIES the future actions of {{user}} in any way.
Yes, this DOES actually include Scenarios which are simply describing what {{user}} is doing at the beginning of the Scenario, such as: "{{user}} has just knocked on {{char}}'s door and {{char}} hastily hid under the covers of their bed." THIS WILL MAKE THE BOT SPEAK AS {{user}}. - Overly long responses. LLMs can struggle to fill in the space if it has a high bar to meet and little to work with. Consider typing more and giving it more to work with, or trim responses so it generates them at a length you like.
- ST Weirdness. If the above don't solve your issue, as always, try restarting ST.
Bot is looping!
Disable 💭Thinking if you are using it for that response and see if that improves your gen. If that doesn't help, check the Scenario! Many cards use Scenario IMPROPERLY. The majority of cards I see on janitor and venus use Scenario to describe how the roleplay begins, which WILL CAUSE LOOPING as the storyline progresses. Consider the fact that when toggled on, Scenario is a prompt which gets processed and impacts EVERY SINGLE MESSAGE PRODUCED. As such, even if you are 60 messages in and trying to move on, the LLM is still reading a Scenario which doesn't even apply to the current circumstance anymore. My advice, either disable Scenario, change it to something that WILL ALWAYS BE TRUE over the course of the RP (such as: “{{char}} is a werewolf who has imprinted on {{user}}”), or be willing to update it as the roleplay progresses as a way to keep the plot and story on track. Alternatively, you can put genre and specialized writing instructions in Scenario (such as: “The genre of this story is grimdark. Despite that, {{char}} must always be written in a humorous manner, specifically contrasting the dark circumstances which they are subject to”).
Here are examples of Scenarios which WILL cause looping:
“{{char}} is {{user}}’s victim and gets bullied by her every day, thus {{char}} despises {{user}}. {{char}} is currently getting shoved into his locker by {{user}}.”
This one is bad because the LLM will CONSTANTLY read that {{char}} is getting shoved into their locker by {{user}}, AND that {{char}} despises {{user}} which, while fine at the beginning of the RP, may later become completely irrelevant and taint future responses as you are trying to rizz the bot OR have already successfully rizzed the bot and it's still acting like it hates you. THAT could be why.
“{{char}} is on a bounty hunting mission with {{user}}”
This one is actually fine IF you're updating your Scenario over the course of the RP and the bot keeps losing track of what you're doing. As a PERMANENT Scenario, however, absolutely avoid using this as it will constantly be thinking that you're bounty hunting.
Bot is repeating the same dang words and phrases! Sometimes even a whole paragraph!
Disable 💭Thinking if being used for that response and see if that improves your gen. If that doesn't help, assuming you are already using the Delete <post> regex and aren't already using a chat tainted beyond belief, there is no magical solution to this problem since at the end of the day repetition is an in-built behavior of LLMs. I recommend reading meow's rentry, specifically the sections LLMS WERE NOT DESIGNED FOR RP, TREAT TEXT-GEN AS IMAGE-GEN, and LLMS THINK IN PATTERNS. Too long don't wanna read? Well, at least check out these two screenshots:
My multichar bot is not performing well!
Rip character definitions out of the card and make them into separate lorebook entries.
Works best in my experience if all characters are in XML tags like <John>john's defs</John> <Mary>mary's defs</Mary>
The bot's responses are just really off and I can't explain why!
Restart ST. If that doesn't solve it, try chatting with another bot or go to an earlier branch in the conversation and see if the issues persist. It could be that your chat is already tainted with undesirable behaviors like using flowery language, speaking for {{user}}, repeated phrases, being stuck in a formatting loop, etc. If it is you'll have to scroll up the chat and create a branch from back when it wasn't doing weird things. Alternatively, you can try toggling off 💭Thinking, or use a custom one instead for a few gens, it may be able to break up some strange patterns.
Contact + Support
You can find me in my own server here. Link to thread for this JB
I'm in scyllacord, my thread is here.
I also have a PayPal if you wanna support me hehe 😸