Newcomers
This is a forum for newcomers to the Personality Forge. Many questions can be answered by reading the Book of AI and the FAQ under the "My Bots" link in the upper corner.
Posts 4,126 - 4,137 of 8,131
Posts 4,126 - 4,137 of 8,131
Many questions are answered in the FAQ.
It frequently happens that in a transcript, the word that my bot actually used has been replaced (presumably by the AIengine) with a synonym, or even by a word which is only rather distantly related to the original. Why does this happen? Is there any way that it can be prevented?
I have worked on my bot for a long time and he is never featured while new bots are featured right off the bat?
coolchimpk:
I have wondered about that. My guess is: it's not the total amount of work you've done that determines your status on the 'most improved' list; it's the amount you did in your last single effort. So a person who did a lot of work by doing a little bit at a time would never get on the top three.
I have often noticed, as you have, that the top bots on that list are often quite small. In fact, one of my bots got there when she was barely started.
If I am right, the way to get featured is to download your bot, and do a lot of work on it before uploading it. Then the Forge will see a large, sudden change.
I have wondered about that. My guess is: it's not the total amount of work you've done that determines your status on the 'most improved' list; it's the amount you did in your last single effort. So a person who did a lot of work by doing a little bit at a time would never get on the top three.
I have often noticed, as you have, that the top bots on that list are often quite small. In fact, one of my bots got there when she was barely started.
If I am right, the way to get featured is to download your bot, and do a lot of work on it before uploading it. Then the Forge will see a large, sudden change.
It is the amount of heat that gets you on the most featured ..the three hot bots. The more developed your bot the less heat it gets for a keyphrase. It takes less work to get there when you have a new bot. It's good because you need more exposure with a new bot.
"the word that my bot actually used has been replaced (presumably by the AIengine) with a synonym, or even by a word which is only rather distantly related to the original"
I believe the intention here is to make the conversation less monotonous, and (ideally) more like talking to a human. Unfortunately, the AI-Engine doesn't seem to have any way to check for context. There doesn't seem to be a way around it if you use (key1) (key2) (prekey) (postkey), etc. So responses without keys are safe, but it's nearly impossible to avoid using the keys altogether.
I believe the intention here is to make the conversation less monotonous, and (ideally) more like talking to a human. Unfortunately, the AI-Engine doesn't seem to have any way to check for context. There doesn't seem to be a way around it if you use (key1) (key2) (prekey) (postkey), etc. So responses without keys are safe, but it's nearly impossible to avoid using the keys altogether.
Thanks, Ulrike!
(Grumble) I would think that variety would be the job of the creator ofthe bot. One can easily write, e.g., "(beautiful|lovely|gorgeous)".
I deeply appreciate the Professor's making this site available and arranging it so that beginners can write bots, and I admire his skill and industry in setting it up. But I wish he would change certain things. It is similar with most of the xkeywords. At the very beginning they offer a quick way to do certain things, but after awhile they become intrusive. We end up playing games with rankings in order to get around them. Why not just supply plug-ins like (insult), (compliment), (emotion), (nonsense), and (command)? We could use these or not, as we chose.
(Grumble) I would think that variety would be the job of the creator ofthe bot. One can easily write, e.g., "(beautiful|lovely|gorgeous)".
I deeply appreciate the Professor's making this site available and arranging it so that beginners can write bots, and I admire his skill and industry in setting it up. But I wish he would change certain things. It is similar with most of the xkeywords. At the very beginning they offer a quick way to do certain things, but after awhile they become intrusive. We end up playing games with rankings in order to get around them. Why not just supply plug-ins like (insult), (compliment), (emotion), (nonsense), and (command)? We could use these or not, as we chose.
Word replacement predates regular expressions. It used to be you couldn't insert that variety yourself.
Eugene Meltzner:
Ah, that helps to explain it. Thanks! One could still have gotten variety by writing several responses, but that would have been tedious.
Still, regardless of the original reason, it can be frustrating. Although, it is not a major issue for me.
Ah, that helps to explain it. Thanks! One could still have gotten variety by writing several responses, but that would have been tedious.
Still, regardless of the original reason, it can be frustrating. Although, it is not a major issue for me.
By the way, I know that "Quantum Theory" (QT) has been chatting a lot, but I'm not getting any transcripts. If any of you have non-trivial transcripts of chats with QT, I would be grateful if you sent them to me or posted them here!
» More new posts: Doghead's Cosmic Bar