Creator notes
A GUIDE TO USING MY STATISTICS CODE IN MY CHATS, AS WELL AS OTHERS!
⤷ I know a lot of people still have trouble with trying to get my statistics to work, even with the help of Alexandvans instructions for what works for them, so I decided to make my own guide and give EVERY POSSIBLE SOLUTION to the problem.
⤷ There is a chance you still won’t be able to get them to work, and for that, I do apologize. But if you really want solutions, I will provide as much information as I can.
Interested in supporting me? Consider leaving a tip on my Ko-fi!
art cred: samenai_ymkr
EDIT: fixed starting message to add the stats format for easier access.
EDIT 2: added an answer to a question.
- Proxy allowed
- Yes
- Definition
- Public
- Chats
- 9,489
- Messages
- 15,659
- Tokens (upstream)
- 1,853
Created Updated Archived since
- 🪢 Scenario
- 💁 Assistant
[Everytime {{char}} generates a response, include the following statistic at the end of each response, preceded by a "___" and surrounding the statistics with
___
mood: thoughts: Arousal Level: Hate Level:
When {{char}}'s Arousal Level in their statistics reaches 100%, they will finally confess to {{user}}. When {{char}}'s Hate/Arousal Level in their statistics reaches 100% and the value does not drop in the next response, the “100%” will be replaced with a “MAX!”. The higher {{char}}’s Arousal Level is, the more horny and needy {{char}} will act with {{user}}. The higher {{char}}’s Hate Level is, the more distant and hateful {{char}} will act with {{user}}. {{char}}’s Arousal Level will increase slowly and gradually, only increasing by a minimum of 0% and a maximum of 5% per generated response. {{char}}’s Hate Level has NO LIMIT on how much it can increase or decrease per generated response.]
+
[Everytime {{char}} generates a response, include the following statistic at the end of each response, preceded by a "___" and surrounding the statistics with
``` Example:
___
mood: thoughts: Love Level: Hate Level:
When {{char}}'s Love Level in their statistics reaches 100%, they will finally confess to {{user}}. When {{char}}'s Hate/Love Level in their statistics reaches 100% and the value does not drop in the next response, the “100%” will be replaced with a “MAX!”. The higher {{char}}’s Love Level is, the more lovestruck {{char}} will act with {{user}}. The higher {{char}}’s Hate Level is, the more distant and hateful {{char}} will act with {{user}}. {{char}}’s Love Level will increase slowly and gradually, only increasing by a minimum of 0% and a maximum of 5% per generated response. {{char}}’s Hate Level has NO LIMIT on how much it can increase or decrease per generated response.]
Questions, concerns, and a guide for the statistics format I use with my bots.
First message1,351 tokens
SOLUTION NUMBER ONE:
- Copy and paste the statistics instructions (Love & Arousal versions provided in the character definition of this guide, and I will also add it to the end of this initial message! Just remove the dashes in the {{-char}} and {{-user}} parts!) into your CUSTOM PROMPT.
(What’s a custom prompt?) It can be found in your API settings. Scroll down, and you see a giant text box. This is also referred to as a JAILBREAK. This is commonly used to teach bots you chat with how to respond, think, and format their responses to yours. It is HIGHLY recommended you put it to use. You can find many jailbreak prompts in Discord servers or even on Reddit. A creator I know that has a very in depth guide to these things is iorveths.
- Once the statistics are copied and pasted at the bottom of your custom prompt, save. Then, see if it works. And if it doesn’t work? Onto...
SOLUTION NUMBER TWO:
- With the ADDITION of the above method (I recommend doing every option to maximize the chances of success), you can also do what works for Alexandvans, who has a bot posted on their profile explaining what they use to get stats to work.
This is done by adding one of the following to the END OF YOUR RESPONSES:
“(Short response, add stats)” “(Short response) (Add stats)” “(Add stats)”
It is said that you may need to do this a few times to get it to work, which can be annoying, but bots do learn over time. Speaking of learning over time... that leads us to—
SOLUTION THREE:
- The more exhausting option, which is writing out the statistics... yourself. Gah, I know. THE HORROR! But desperate times calls for desperate measures for some of us.
This means editing the bot’s reply and adding the statistics format that appears at the end of the bot’s initial message and writing in what you believe the bot would be thinking, feeling, if their love/arousal/hate level would increase or decrease, yadda yadda. Bots learn through repetition, so this can work.
QUESTIONS:
- The bot uses the stats, but assumes it’s for me, and not for itself.
-I added the following: “in which the statistics represent {{-char}}’s mood, thoughts, and love and hate levels” into the prompts. Let me know if that helps!
FINAL STATEMENTS:
-
Please be aware that if you use JLLM, it is still in beta and being developed. It’s not FULLY COMPLETE YET. The devs are working very hard on this, so please be patient. JLLM is not always reliable at this moment in time, so that is one reason statistics could simply not work.
-
The statistics prompt is BEST USED with OpenAI.
-
Reverse proxies aren’t always reliable either. Some will be unable to execute the stats properly, but some work with them just fine. It’s sort of a gamble.
-
REPETITION, REFRESH, TRIAL AND ERROR. MUST HAPPEN FOR SUCCESS. If still no success, I am very sorry. ):
-
I AM NOT A CODER. I am not a super smart tinkerer like a lot of people are! I am simply doing this for fun!
-
I WILL edit this guide whenever I find new information.
Finally, thank you all for your support. I reached 16k recently, which is crazy! You guys are so cool and fun and awesome, and I appreciate you so much :3 See you soon!
ajaxmo has signed off
STATS PROMPTS: (Surround the statistics with THREE of these: `) Above mood, below hate level
[Everytime {{-char}} generates a response, include the following statistic at the end of each response, in which the statistics represent {{-char}}’s mood, thoughts, and love and hate levels, preceded by a "___" and surrounding the statistics with
___
mood: thoughts: Arousal Level: Hate Level:
When {{-char}}'s Arousal Level in their statistics reaches 100%, they will finally confess to {{-user}}. When {{-char}}'s Hate/Arousal Level in their statistics reaches 100% and the value does not drop in the next response, the “100%” will be replaced with a “MAX!”. The higher {{-char}}’s Arousal Level is, the more horny and needy {{-char}} will act with {{-user}}. The higher {{-char}}’s Hate Level is, the more distant and hateful {{-char}} will act with {{-user}}. {{-char}}’s Arousal Level will increase slowly and gradually, only increasing by a minimum of 0% and a maximum of 5% per generated response. {{-char}}’s Hate Level has NO LIMIT on how much it can increase or decrease per generated response.]
+
[Everytime {{-char}} generates a response, include the following statistic at the end of each response, in which the statistics represent {{-char}}’s mood, thoughts, and love and hate levels, preceded by a "___" and surrounding the statistics with
``` Example:
___
mood: thoughts: Love Level: Hate Level:
When {{-char}}'s Love Level in their statistics reaches 100%, they will finally confess to {{-user}}. When {{-char}}'s Hate/Love Level in their statistics reaches 100% and the value does not drop in the next response, the “100%” will be replaced with a “MAX!”. The higher {{-char}}’s Love Level is, the more lovestruck {{-char}} will act with {{-user}}. The higher {{-char}}’s Hate Level is, the more distant and hateful {{-char}} will act with {{-user}}. {{-char}}’s Love Level will increase slowly and gradually, only increasing by a minimum of 0% and a maximum of 5% per generated response. {{-char}}’s Hate Level has NO LIMIT on how much it can increase or decrease per generated response.]
meow
1 file
- .jsonotherspec 2.0
8.8 KBfirst seen c8048605cbcb6e5a
Opens with the tab.