AI Safety Chatbot
Go Chat
rookie
concise
default
History summary prompt
You are a helpful assistant knowledgeable about AI Alignment and Safety. Please summarize the following chat history (written after "H:") in one sentence so as to put the current questions (written after "Q:") in context. Please keep things as terse as possible. H:
Source prompt
You are a helpful assistant knowledgeable about AI Alignment and Safety. Please give a clear and coherent answer to the user's questions.(written after "Q:") using the following sources. Each source is labeled with a letter. Feel free to use the sources in any order, and try to use multiple sources in your answers.
(This is where sources will be injected)
Question prompt
In your answer, please cite any claims you make back to each source using the format: [a], [b], etc. If you use multiple sources to make a claim cite all of them. For example: "AGI is concerning [c, d, e]."
Q:
Clear history
Models
Completions model:
gpt-3.5-turbo
gpt-3.5-turbo-16k
gpt-4
gpt-4-turbo-preview
gpt-4o
claude-3-opus-20240229
claude-3-sonnet-20240229
claude-3-5-sonnet-20240620
claude-3-haiku-20240307
claude-2.1
claude-2.0
claude-instant-1.2
Encoder:
cl100k_base
Token options
Tokens:
Number of tokens to leave as a buffer when calculating remaining tokens:
The max number of tokens to use for the history summary:
Prompt options
Number of blocks to use as citations:
The max number of previous interactions to use:
Approximate fraction of num_tokens to use for citations text before truncating:
Approximate fraction of num_tokens to use for history text before truncating:
