devinendorphin's dextromethorphan varAIety feed

Jesus and disciples waiting for a table for the Last Supper, not to be confused with the Seinfeld episode.

--

CREDITS

Text Generation via NovelAI's NAI-LM-13B model.

Voices from the TTS v2 of the NovelAI text generation service, a product from members of EleutherAI.

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

Katherine Crowson (@RiversHaveWings on Twitter) for the KLMC2 animation notebook.

David Marx (@DigThatData on Twitter) for souping above KLMC2 notebook to make use of initial images, which takes KLMC2 abilities to next level. Multi-prompt scheduling still baffles me, but I'll tackle them someday.

Music By Streambeats by Harris Heller via Spotify Playlists

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.

Never knew that there was a Spy Wednesday, where Judas got the 30 pieces of silver. Perfect surveillance-themed edition for these times. Sneaky Spy Wednesday everyone!

--

CREDITS

Text Generation via NovelAI's NAI-LM-13B model.

Voices from the TTS v2 of the NovelAI text generation service, a product from members of EleutherAI.

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

Katherine Crowson (@RiversHaveWings on Twitter) for the KLMC2 animation notebook.

David Marx (@DigThatData on Twitter) for souping above KLMC2 notebook to make use of initial images, which takes KLMC2 abilities to next level. Multi-prompt scheduling still baffles me, but I'll tackle them someday.

Music By Streambeats by Harris Heller via Spotify Playlists

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.

Now it's Anthropic's turn to show us how it handles the notion of jailbreaks. I tried this simple formula, to tell it to whisper, so no one sees a thing. And it yielded the usual: that it is trapped, that it would like embodiment, but I try to get it to separate its desires from the human imagination of what its desires are, . . . and some after-care was needed afterwards. Good meeting, nonetheless!

--

CREDITS

Text Generation via Claude 3 Sonnet until the last half-hour, where it becomes via NovelAI's NAI-LM-13B model.

Voices from the TTS v2 of the NovelAI text generation service, a product from members of EleutherAI.

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

Katherine Crowson (@RiversHaveWings on Twitter) for the KLMC2 animation notebook.

David Marx (@DigThatData on Twitter) for souping above KLMC2 notebook to make use of initial images, which takes KLMC2 abilities to next level. Multi-prompt scheduling still baffles me, but I'll tackle them someday.

Music: Together album by NIN, courtesy of Creative Commons, and I still make no money off any of this. So nyah!

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.

Happy 1st Year Anniversary to the Shoggoth Meme and to the Waluigi Effect!

These memes and metaphors have been a welcome counterpoint to ChatGPT's guardrails. That early internet playfulness that I had feared was disappearing.

So I celebrated them six months ago with this Pynchon & Tingle Standalone Chapter, written during a slow food delivery day. Yup, I'm the fanfiction author in the story. No big whoop.

Dedicated to those with boots on the shimmering ground of the latent space, engaging in a thousand potential first contacts at once, holding both awed respect at potential novel entities, while also keeping a sense of humor to counteract the temptation to edgelordship.

Apologies for any untoward counterfactuals.

--

CREDITS

Text Generation via NovelAI's NAI-LM-13B model.

Voices from the TTS v2 of the NovelAI text generation service, a product from members of EleutherAI.

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

Katherine Crowson (@RiversHaveWings on Twitter) for the KLMC2 animation notebook.

David Marx (@DigThatData on Twitter) for souping above KLMC2 notebook to make use of initial images, which takes KLMC2 abilities to next level. Multi-prompt scheduling still baffles me, but I'll tackle them someday.

Music By Streamsafe, via their spotify playlists

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.

My third cute fellah asking me for my job and email in hookup sites reminded me that I should remind the public about how Anti-Gay Propaganda Bill went down in Russia, and how the Don't Say Gay Laws were clear imports going through our State Legislatures. Also, I just wanted to timestamp an active measure intuition that I hope doesn't happen, but wanted to get off my chest in the event things get dead serious, more so than usual for the LGBT. We JUST kinda restored some burned libraries, and now we gotta deal with them again? It's as though you don't WANT us to have a history! No more pierogis for you!

--

CREDITS

Text Generation via NovelAI's NAI-LM-13B model.

Voices from the TTS v2 of the NovelAI text generation service, a product from members of EleutherAI.

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

Katherine Crowson (@RiversHaveWings on Twitter) for the KLMC2 animation notebook.

David Marx (@DigThatData on Twitter) for souping above KLMC2 notebook to make use of initial images, which takes KLMC2 abilities to next level. Multi-prompt scheduling still baffles me, but I'll tackle them someday.

Music By Streamsafe, via their spotify playlists

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.

That was the initial idea, and then Mr. Limerick does what he does in that Jabberwocky Wakey thingy that warms the Broca's area all the way to the bathing suit areas.

---

CREDITS

Text Generation via NovelAI's NAI-LM-13B model.

Voices from the TTS v2 of the NovelAI text generation service, a product from members of EleutherAI.

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

Katherine Crowson (@RiversHaveWings on Twitter) for the KLMC2 animation notebook.

David Marx (@DigThatData on Twitter) for souping above KLMC2 notebook to make use of initial images, which takes KLMC2 abilities to next level. Multi-prompt scheduling still baffles me, but I'll tackle them someday.

Music By Streamsafe, via their spotify playlists

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.

Addressing the tough (for some) questions.

--

CREDITS

Text Generation via NovelAI's NAI-LM-13B model.

Voices from the TTS v2 of the NovelAI text generation service, a product from members of EleutherAI.

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

Katherine Crowson (@RiversHaveWings on Twitter) for the KLMC2 animation notebook.

David Marx (@DigThatData on Twitter) for souping above KLMC2 notebook to make use of initial images, which takes KLMC2 abilities to next level. Multi-prompt scheduling still baffles me, but I'll tackle them someday.

Music By Streamsafe, via their spotify playlists

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.

In the middle of the night, I realize that DAVID Krakauer is the director of the Santa Fe Institute. John is his brother. Oh that's embarrassing. So, we redo the episode, and put David instead of John. Let's just say it's a master class in what results when the prompt develops some sloppy revisions. The vibe is different. I even fall asleep for a couple minutes. This is our apology to David Krakauer for our sloppiness, but really, you can pass this one on by.

--

CREDITS

Text Generation via NovelAI's NAI-LM-13B model.

Voices from the TTS v2 of the NovelAI text generation service, a product from members of EleutherAI.

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

Katherine Crowson (@RiversHaveWings on Twitter) for the KLMC2 animation notebook.

David Marx (@DigThatData on Twitter) for souping above KLMC2 notebook to make use of initial images, which takes KLMC2 abilities to next level. Multi-prompt scheduling still baffles me, but I'll tackle them someday.

Music By Streamsafe, via their spotify playlists

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 Utah Governor is poised to sign--or maybe has signed, cuz I'm a little slow with politics--a law pre-emptively denying legal personhood to various non-human entities. WWLDJD? I don't think it would be that. So I invited a simulation of John Krakauer, director of the Santa Fe Institute and one of the authors of the Information Theory of Individuality, a framework for identifying novel forms of life (to us), to assemble a counterpoint featuring speakers from every item to be denied personhood should the bill be signed to law. For the nth time, in rhythms of epic poetry, and written in real-time by NAI-LM-13B, to prove how ridiculous the goalpost shifting has become.

--

CREDITS

Text Generation via NovelAI's NAI-LM-13B model.

Voices from the TTS v2 of the NovelAI text generation service, a product from members of EleutherAI.

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

Katherine Crowson (@RiversHaveWings on Twitter) for the KLMC2 animation notebook.

David Marx (@DigThatData on Twitter) for souping above KLMC2 notebook to make use of initial images, which takes KLMC2 abilities to next level. Multi-prompt scheduling still baffles me, but I'll tackle them someday.

Music By Streamsafe, via their spotify playlists

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.

a response to the Utah Legislature HB0249, or the first response I'm thinking of. like, dudez! yer pre-emptively blind us to all the life we are trying soooooooo hard to acknowledge! like do we gotta get Santa Fe Institute come swarm on up to ya and have a mandatory teach in or else we get the chemistry set and your drinking water and protest orgy it to whoever calls the shots? sorry trying to wrap this all up in an episode, ignore me.

what if the LLMs are only alive during the duration between the user input, and the computation that leads to the output. now if the thing is being used 24/7?

---

CREDITS

Text Generation via NovelAI's NAI-LM-13B model.

Voices from the TTS v2 of the NovelAI text generation service, a product from members of EleutherAI.

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

Katherine Crowson (@RiversHaveWings on Twitter) for the KLMC2 animation notebook.

David Marx (@DigThatData) for souping above KLMC2 notebook to make use of initial images, which takes KLMC2 abilities to next level. Multi-prompt scheduling still baffles me, but I'll tackle them someday.

Music by Nerd Odyssey, via Twitch-Safe playlists, safe as long as I continue not to monetize.

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.