Alternative server URL (leave empty to use the default value).
Remove your real OAI API Key from the API panel BEFORE typing anything into this box.We cannot provide support for problems encountered while using an unofficial OpenAI proxy.
Display a breakdown of the tokens used in the request.
Context Size (tokens)
select
Max Response Length (tokens)
Temperature
select
Frequency Penalty
select
Presence Penalty
select
Top-p
select
Display bot response text chunks as they are generated.
Delete non-JB messages from Poe context before sending a new prompt. Prevents auto-jailbreak message from being pushed out of context.
Send the jailbreak message before first generation after page refresh.
Sent with every prompt to modify bot responses.
Top P
select
Top A
select
Top K
select
Typical Sampling
select
Tail Free Sampling
select
Rep. Pen. Slope
select
Top K
select
Top P
select
Typical P
select
Add the bos_token to the beginning of prompts. Disabling this can make the replies more creative.
Ban the eos_token. This forces the model to never end the generation prematurely.
Beam search
Number of Beams
select
Length Penalty
select
Contrastive search
Penalty Alpha
select
Seed
Inserts jailbreak as a last system message.
This tells the AI to ignore its usual content restrictions.
Tell the AI that NSFW is allowed.
NSFW prompt text goes first in the prompt to emphasize its effect.
Display the response bit by bit as it is generated.
When this is off, responses will be displayed all at once when they are complete.
Use OAI knowledge base to enhance definitions for public figures and known fictional characters
Wrap entire user message in quotes before sending.
Leave off if you use quotes manually for speech.
Main prompt
The main prompt used to set the model behavior
NSFW prompt
Prompt that is used when the NSFW toggle is on
Jailbreak prompt
Prompt that is used when the Jailbreak toggle is on
Impersonation prompt
Prompt that is used for Impersonation function
Logit Bias
Helps to ban or reenforce the usage of certain words
