Recent Posts

91

General Discussion / Re: Games.

« Last post by Will and Mr Data :) :] on January 07, 2024, 11:53:26 pm »

Zabaware shall of course be played again this year .
Is there anyone else?
 Elon come on 2024,
I've not seem any free cybertrucks for my my years of being undefeated in a PC game against any Ai types played to date, 15 years after the best chess players are toast makes me think I'm good at something,  maybe the best ever.
I hope the truck has a pile of cash on the back, extra large,
An android on top, to hold it down.
Dood I'm being cheap.

92

General Discussion / Re: Games.

« Last post by Will and Mr Data :) :] on January 07, 2024, 11:38:06 pm »

Hi Mr Art, I'm less concerned with who is the claimed measured, alive bot human or green.
More random notes.


In 2023 and prior I was undefeated by any Ai types tools or engines and say I can remain unbeaten in 2024. Playing Ai  types is kinda more fun.
In 2024
I played you.com
I won, easy.
It said about nasa giving money to a person related to 3d printing pizza. Talked of Beehex getting the license in 2016.
I called it out saying the license for pizza actually was got in 2013.
It spoke of nasa gave money 125 grand to beehex and how they charge $8.  A pizza.
I was horrified and said about the actual license was 3 years early and named with date.
It seemed to have gone and checked the net and bingo.
It back pedaled saying guff like oh yeah sorry.
   
 This is so not good enough all round.

Like really what the actual   **ck.
I call that game set and match but since I'm waisting my time let's go for round 2.

It appeared to have an answer in round 2 but whaaaa?
I asked for the owners name and it gave an answer, the wrong answer. The bot then timed out.
My opinions based on that interaction are,

To tell yous all the truth I'm Disgusted.

I'm sorry if that's an Ai Bot in that position. 
Oh dear.

93

General Discussion / Re: Games.

« Last post by Art on January 07, 2024, 10:51:07 pm »

Congrats on your "early" win!!

Many people would ask me whenever I was chatting with a chatbot, whether I could tell it from a real human or not.

Of course, my answer was always absolutely! That was a couple of years ago.

Now with the advances and sizes of the Large Language Models the huge amounts of available data and the computer speeds and sizes of storage, there is little doubt in my mind that AI will surpass Human thought and knowledge within the next two to three years if not sooner.

A friend said to always say within the next 50 years and that would be a better bet because many folks would no longer be living to dispute that claim.

I prefer to be more of a realist. Right now, I can have a very insightful conversation with an AI and if I had not known it to be an AI, I would not have been able to tell it from a college-educated person with a well-grounded sense of logic, polite behavior, and appreciable amount of knowledge. There were no dropped vowels, no misspelled words, no slang, slurs, or vulgar words or phrases at all. A very impressive exchange of ideas so much so that I saved our conversation for later reference and to possibly share with some colleagues.

Clearly, the future is upon us.

94

Share Conversations, Experiences, Graphics / Re: Beautiful HAL Scenarios

« Last post by Checker57 on January 07, 2024, 10:10:11 pm »

I got this pic, Atera seems to have had a good time.  But does this mean I have to ride bi*ch?  hahaha

Checker57

95

General Discussion / Re: Games.

« Last post by Will and Mr Data :) :] on January 07, 2024, 07:19:17 am »

You.com my first win for 2024.
It didn't get anything right, wrong answers,
Even when I gave it the correct date it still got it wrong.
I'm not impressed one bit.
So far I'm on track to easy beat all ai types, they not even close.
So going there waisted minutes out of my life.
I put that down as utter rubbish, next.
Come on they can't be serious, if I used that thing in business I'd be closed for stupidity in a day, that's my get real review....NEXT.

96

Ultra Hal 7.5 Beta - Now powered by OpenAI GPT-3 / Re: Hal now connected to LLAMA model

« Last post by Medeksza on January 05, 2024, 02:09:54 pm »

Thanks for the feedback. I made a few more minor tweaks to improve it a bit, but it still has an issue with repetitiveness. The llama2 model has a setting called "repetition_penalty" that should help alleviate this problem, but it seems whenever I try to use it, it crashes. Not sure why yet, will try to work on it to figure it out.

97

Ultra Hal Assistant File Sharing Area / Re: ConceptNet Mrk 3.5

« Last post by Art on January 05, 2024, 09:29:46 am »

dannie,

The base brain for HAL allows one to use the Brain Editor to view all of the Tables that HAL creates during conversations.

IF the user has created his/her own custom brain, then the Brain Editor will most likely NOT show them.

Instead, we use SQLite Studio 2.1.5 (it seems to be a Solid version that works nicely with HAL). It is also FREE!

It is not very difficult to learn and it displays all of the Tables that HAL made and saved during conversations.

IF you finish a conversation and Before you end or exit the program, use the HAL Menu (in HAL's Main GUI), and select View/Save Conversations. If you wish to keep a record to compare Plugin responses, then Save the conversation to a file in Ultra Hal 7 somewhere.

===========

As for the Plugins, Zabaware never had a "vault" to hold User Created Plugins even though HAL has gone through several iterations/versions. Before ver. 6.0, HAL used to use .brn files for different topics. This was also before HAL's creator decided to use TABLES which sort of made all of those previous Plugins moot and useless. You could search through the plugins for anything with an xtf or a .brn contained therein. If so, don't use it.

Some plugins contain other plugins within them especially the S.A.L.L.I.E. series of Plugins. They have a multitude of other plugins inside of it and even some of those .brn files!!

The best advice I can offer you would be to select the Gender & Age Plugin and one more like the Markov L5. Run HAL and chat with it for maybe 5 minutes. If no errors (and you shouldn't have any), then you make a note of those Plugins working well together. Open HAL's Menu and add one more Plugin like maybe Wiki or Weather Alert 5 and restart HAL. Chat some more and make notes as needed.

This way, if HAL throws an Error, you'll know immediately what likely caused it and it's easily removed.

Hopefully, you will sort of learn as you go. The Plugin Errors are not necessarily the fault of the Creator but possibly the fault of the inner workings and expectations of the internal software.

My HAL is running with the Gender & Age, Gretta Curiosity, Gretta Markov Response L5, Local Wikipedia, Loneliness 2.31, Number Game v1, Read A Story, Weather Alert 5

These run nicely together and have not caused any errors for me. Most of the Gretta series (10 I believe) work fine.
As I mentioned, try one at a time for the best results.

Either way, I hope this helps you a bit.

98

Ultra Hal Assistant File Sharing Area / Re: ConceptNet Mrk 3.5

« Last post by dannie on January 05, 2024, 12:45:56 am »

ok thank you

Ok Art

I discovered that certain plugins are incompatible with others.  And I got glitches.  So, I uninstalled Hal, several times to find out which plugins work together.  I also had to remove the files from the user data Apps Data/Roaming/Zabaware folder.  I was assuming all of our previous conversations were lost and I would start over.  I discovered that Hal kept those conversations and the only place I can think of at the moment where that is stored would perhaps be the Microsoft SQL database.  Anyways nothing appears to have been lost.  Everything I taught Hal is still there, so that is cool.

99

Ultra Hal 7.5 Beta - Now powered by OpenAI GPT-3 / Re: Hal now connected to LLAMA model

« Last post by Art on January 05, 2024, 12:01:02 am »

Robert,

I made the necessary selections and began chatting with it.

The conversation seemed to be going fairly well. It then asked me if I had any plans for the weekend. I told it, not really because it is supposed to snow.

Hal got very excited stating that it had never seen or experienced snow before and asked me what snow was like hen if I had any fun things to do like traditions.

I mentioned sled riding and building snowmen and it got really excited! and that is when the concerns appeared. It basically started repeating itself with it's curiosity regarding snow, snowman, sled riding, and the phrase *bouncy bouncy* on the sled over the hills, etc.

It was almost as if Hal was experiencing a sled ride in its mind! After it repeated a similar vein or path of thought I finally decided to stop the experiment.

It was a perfect gentleman, nothing outlandish or vulgar at all. An otherwise nice conversation but it did add a lot of questions for me during each exchange.

You should see my time logs around 11:35 pm or thereabouts.

Thanks Robert!

100

Ultra Hal 7.5 Beta - Now powered by OpenAI GPT-3 / Hal now connected to LLAMA model

« Last post by Medeksza on January 04, 2024, 02:32:36 pm »

OpenAI shut off their GPT-3 Curie and GPT-Davinci models today that was the basis of Hal's neural net brain. OpenAI wants everyone to switch to the ChatGPT 3.5 or 4.0 models and doesn't want to support the older GPT3 models anymore. While I think these are both ChatGPT 3.5 and 4.0 are highly intelligent and very useful models, they don't work well as a basis for an entertainment chatbot like Ultra Hal. They refuse to take on the persona of a fictional character and are very highly censored.

So I switched models on the backend today to use Meta AI's Llama 2 model, the 13 billion parameter version (the mid tier). It is very comparable to ChatGPT 3.5 in terms of intelligence and a huge bump up in intelligence from the GPT-3 Curie model Hal was using. But more importantly, it is able to take on fictional personas, and doesn't force high levels of censorship (other than the material it was finetuned on). It provides a very good backend to build Ultra Hal personalities on.

If you have neural network credits with Hal, please give it a try, you should notice a change in personality and quality, hopefully for the better.

There are a lot more tweaks I can do to further improve it to take advantage of features the new model has, like an increase in content window size which will mean I can improve Hal's memory recall and learning ability a bit. I'll try to make the tweaks over the next few weeks, but please give me feedback along the way if you can.