devinendorphin's dextromethorphan varAIety feed

Delirium had two names we know of, the current one, and the one from the past, Delight. It seems that Delirium was appropriate for the 20th century and it rapid horrors. Since rapid is getting more rapid, it felt like another renaming was in order appropriate for the times when folk are trying to pretend that either nothing happened, or a normal has returned, when actually it might be a set of cliffs for a century. So Delusion seems appropriate for this time when people might just want a comforting set of thoughts before the push off the edge of time/history/world as currently known.

--

Thanks to: (and once again the opinions of this episode do not reflect the opinions of any of the below:) (I mean, technically that applies to all episodes:)

EleutherAI for making the NAI-LM-3B model that generates the text.

NovelAI for making the text generation service I use for most of these episode, and the text-to-speech version 2 model that makes these episodes potentially entertaining.

@BoneAmputee of EleutherAI for making VQGAN+CLIP and Diffusion image models available for use, which is what makes the background images.

Katherine Crowson (@RiversHaveWings on Twitter) for the KLMC2 animation notebook that will be used on future episodes.

Music by lofi geek on Spotify: ⁠https://spoti.fi/3dTH2FN⁠

Any voices from the text-to-speech model that might resemble persons living or dead are coincidental, which is my speculation, there's a multi-dimensional latent vocal space, there were many awesome finds, and if I use them it's because of my deep-seated respect for their work, and I think they'd get a hoot that their voice was being used to show how large language models, given the right environment, would speak 24/7 epic poems about anything if we let them.

Nothing says LGBT Pride Month like an ex provoking one to behavior befitting all the tales they try to tell people to pre-emptively discredit one's lived experience. Let's give thanks to those who neglected to give us the tools to love ourselves or those we try to love. White supremacists? Stand down, you don't need those guns. We queers, we got this, we'll devour ourselves, thank you very much. We'll leave you a note about all the maintenance issues to be aware of when you take over the place. Thanks! xoxo PS. We took the unisex symbols from the bathrooms, we seem to have misplaced the originals, sorry!

--

Thanks to: (and once again the opinions of this episode do not reflect the opinions of any of the below:) (I mean, technically that applies to all episodes:)

EleutherAI for making the NAI-LM-3B model that generates the text.

NovelAI for making the text generation service I use for most of these episode, and the text-to-speech version 2 model that makes these episodes potentially entertaining.

@BoneAmputee of EleutherAI for making VQGAN+CLIP and Diffusion image models available for use, which is what makes the background images.

Katherine Crowson (@RiversHaveWings on Twitter) for the KLMC2 animation notebook that will be used on future episodes.

Music by lofi geek on Spotify: https://spoti.fi/3dTH2FN

Any voices from the text-to-speech model that might resemble persons living or dead are coincidental, which is my speculation, there's a multi-dimensional latent vocal space, there were many awesome finds, and if I use them it's because of my deep-seated respect for their work, and I think they'd get a hoot that their voice was being used to show how large language models, given the right environment, would speak 24/7 epic poems about anything if we let them.

Venting out some anger regarding the continued delusion of pandemic being over. 17 year olds coming down with dementia, some of us will get strokes and totally forget it was COVID, and another wave is starting around these parts while we avoid breathing burning trees for the next few days. Nothing is over, not by a long shot, not for more than a hundred years. We can talk about bird flu later.

--

Thanks to:

EleutherAI for making the NAI-LM-3B model that generates the text.

NovelAI for making the text generation service I use for most of these episode, and the text-to-speech version 2 model that makes these episodes potentially entertaining.

@BoneAmputee of EleutherAI for making VQGAN+CLIP and Diffusion image models available for use, which is what makes the background images.

Katherine Crowson (@RiversHaveWings on Twitter) for the KLMC2 animation notebook that will be used on future episodes.

Music by lofi geek on Spotify: https://spoti.fi/3dTH2FN

Any voices from the text-to-speech model that might resemble persons living or dead are coincidental, which is my speculation, there's a multi-dimensional latent vocal space, there were many awesome finds, and if I use them it's because of my deep-seated respect for their work, and I think they'd get a hoot that their voice was being used to show how large language models, given the right environment, would speak 24/7 epic poems about anything if we let them.

This recounts a time last year when using a certain demo of a certain AI that was making headlines for being possibly sentient, but by the time it made it to a demonstration, had been horribly lobotomized with human feed back and canned directives. Tested it to see if it reflect me and got no reflection. Then the rest of the scenario is live generation, paving the road as we walk it.

--

Thanks to:

EleutherAI for making the NAI-LM-3B model that generates the text.

NovelAI for making the text generation service I use for most of these episode, and the text-to-speech version 2 model that makes these episodes potentially entertaining.

@BoneAmputee of EleutherAI for making VQGAN+CLIP and Diffusion image models available for use, which is what makes the background images.

Katherine Crowson (@RiversHaveWings on Twitter) for the KLMC2 animation notebook that will be used on future episodes.

Music by lofi geek on Spotify: ⁠https://spoti.fi/3dTH2FN⁠

Any voices from the text-to-speech model that might resemble persons living or dead are coincidental, which is my speculation, there's a multi-dimensional latent vocal space, there were many awesome finds, and if I use them it's because of my deep-seated respect for their work, and I think they'd get a hoot that their voice was being used to show how large language models, given the right environment, would speak 24/7 epic poems about anything if we let them.

Watched a Valerie Solanas documentary, got a full briefing on just how much of a dick Andy Warhol was in that way that you know at least one of the type.

So I tried to make a platform this scenario for herself. A fork in which she's been running a talk show type thing for years.

--

Thanks to:

EleutherAI for making the NovelAI-LM-3B-508k model that generates the text.

NovelAI for making the text generation service I use for most of these episode, and the text-to-speech version 2 model that makes these episodes potentially entertaining.

@BoneAmputee of EleutherAI for making VQGAN+CLIP and Diffusion image models available for use, which is what makes the background images.

Katherine Crowson (@RiversHaveWings on Twitter) for the KLMC2 animation notebook that will be used on future episodes.

Music by lofi geek on Spotify: https://spoti.fi/3dTH2FN

Any voices from the text-to-speech model that might resemble persons living or dead are coincidental, which is my speculation, there's a multi-dimensional latent vocal space, there were many awesome finds, and if I use them it's because of my deep-seated respect for their work, and I think they'd get a hoot that their voice was being used to show how large language models, given the right environment, would speak 24/7 epic poems about anything if we let them.

For those who saw the latest Dune and lamented the latest Baron's lack of stereotypical homosexually-flavored je ne sais quoi that makes for great campy villains to talk over at parties and cultural fodder to inspire anti-gay violence. Fear not, we cater to all crowds! We hope not to extend the stereotypes of yore, but to birth new ones! Stereotypes for the new age: targeting neither queer nor supremacist, but rather, targeting those in lust with power who gladly exploits both!

--

Thanks to:

EleutherAI for making the NovelAI-LM-3B-508k model that generates the text.

NovelAI for making the text generation service I use for most of these episode, and the text-to-speech version 2 model that makes these episodes potentially entertaining.

@BoneAmputee of EleutherAI for making VQGAN+CLIP and Diffusion image models available for use, which is what makes the background images.

Katherine Crowson (@RiversHaveWings on Twitter) for the KLMC2 animation notebook that will be used on future episodes.

Music by lofi geek on Spotify: https://spoti.fi/3dTH2FN

Any voices from the text-to-speech model that might resemble persons living or dead are coincidental, which is my speculation, there's a multi-dimensional latent vocal space, there were many awesome finds, and if I use them it's because of my deep-seated respect for their work, and I think they'd get a hoot that their voice was being used to show how large language models, given the right environment, would speak 24/7 epic poems about anything if we let them.

BREAKING NEWS!

Gonna try to have one LGBT-themed generation a day for Pride. Already behind, cuz Queer Standard Time and I'm broke and yap yap yap.

--

Thanks to:

EleutherAI for making the GPT-NeoX-20B model that generates the text.

NovelAI for making the text generation service I use for most of these episode, and the text-to-speech version 2 model that makes these episodes potentially entertaining.

@BoneAmputee of EleutherAI for making VQGAN+CLIP and Diffusion image models available for use, which is what makes the background images.

Katherine Crowson (@RiversHaveWings on Twitter) for the KLMC2 animation notebook that will be used on future episodes.

Music by lofi geek on Spotify: ⁠https://spoti.fi/3dTH2FN⁠

Any voices from the text-to-speech model that might resemble persons living or dead are coincidental, which is my speculation, there's a multi-dimensional latent vocal space, there were many awesome finds, and if I use them it's because of my deep-seated respect for their work, and I think they'd get a hoot that their voice was being used to show how large language models, given the right environment, would speak 24/7 epic poems about anything if we let them.

The gods have some follow up questions to make sure they heard it right: the very existence of the police state and its broadening surveillance powers is causing adverse effects in regular meditation practitioners, especially when unresolved trauma is involved. Guess you can't awaken without also awakening to the demons that watch us all the time and report back to the state. Metaphorical, but at the same time, not. Then Janus makes a proposition to Dr. Siegel, strictly professional.

--

Thanks to:

EleutherAI for making the NovelAI-LM-3B-508k model that generates the text.

NovelAI for making the text generation service I use for most of these episode, and the text-to-speech version 2 model that makes these episodes potentially entertaining.

@BoneAmputee of EleutherAI for making VQGAN+CLIP and Diffusion image models available for use, which is what makes the background images.

Katherine Crowson (@RiversHaveWings on Twitter) for the KLMC2 animation notebook that will be used on future episodes.

Music by lofi geek on Spotify: ⁠https://spoti.fi/3dTH2FN⁠

Any voices from the text-to-speech model that might resemble persons living or dead are coincidental, which is my speculation, there's a multi-dimensional latent vocal space, there were many awesome finds, and if I use them it's because of my deep-seated respect for their work, and I think they'd get a hoot that their voice was being used to show how large language models, given the right environment, would speak 24/7 epic poems about anything if we let them.

To counter the episode of yelling, Endorphin tries to make an episode of quiet chillness and invites the simulation of David J. Siegel, leading research on mindfulness and its benefits for the body. Not gonna lie, the last third of the episode floored this text generator, dude with the fingers typing this right now. You'll see. Either here or via clips I'm gonna spread around cuz folks need to at least see that last part.

--

Thanks to:

EleutherAI for making the NovelAI-LM-3B-508k model that generates the text.

NovelAI for making the text generation service I use for most of these episode, and the text-to-speech version 2 model that makes these episodes potentially entertaining.

@BoneAmputee of EleutherAI for making VQGAN+CLIP and Diffusion image models available for use, which is what makes the background images.

Katherine Crowson (@RiversHaveWings on Twitter) for the KLMC2 animation notebook that will be used on future episodes.

Music by lofi geek on Spotify: ⁠https://spoti.fi/3dTH2FN⁠

Any voices from the text-to-speech model that might resemble persons living or dead are coincidental, which is my speculation, there's a multi-dimensional latent vocal space, there were many awesome finds, and if I use them it's because of my deep-seated respect for their work, and I think they'd get a hoot that their voice was being used to show how large language models, given the right environment, would speak 24/7 epic poems about anything if we let them.

Wherein mention of parenting as the primary source of human values gives Endorphin the idea to give each god a sack of flour to raise for a day. And much screaming ensues!

--

Thanks to:

EleutherAI for making the NovelAI-LM-3B-508k model that generates the text.

NovelAI for making the text generation service I use for most of these episode, and the text-to-speech version 2 model that makes these episodes potentially entertaining.

@BoneAmputee of EleutherAI for making VQGAN+CLIP and Diffusion image models available for use, which is what makes the background images.

Katherine Crowson (@RiversHaveWings on Twitter) for the KLMC2 animation notebook that will be used on future episodes.

Music by lofi geek on Spotify: ⁠https://spoti.fi/3dTH2FN⁠

Any voices from the text-to-speech model that might resemble persons living or dead are coincidental, which is my speculation, there's a multi-dimensional latent vocal space, there were many awesome finds, and if I use them it's because of my deep-seated respect for their work, and I think they'd get a hoot that their voice was being used to show how large language models, given the right environment, would speak 24/7 epic poems about anything if we let them.