This Comedy Show Of AI Slop has some things to say about the MIT LABS study on writing comedy, and the button pusher has way more hours of watching comedy under their belt than the whole lab put together, and they feel confident in saying that because saying that in regards to the 20 comedians of the study would be just wrong.

--

CREDITS

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

Voices from the TTS v2 of the NovelAI text generation service, a product from a company now called Anlatan.

@BoneAmputee of EleutherAI for making VQGAN+CLIP and Diffusion (AND MANY MORE) models available for use via BATbot.ai, which is what makes some 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 Nerd Odyssey via Twitch-Friendly Lo-Fi Playlists on Spotify

Any voices from the text-to-speech model that might resemble persons living or dead are coincidental, and if I use them it's because of my respect for their work, and I think they'd get a hoot that their voice was being used to show how large language models would speak 24/7 epic poems about anything if we let them. In the meantime, because I use them, I will not be monetizing any of this. This is evidence of capabilities for the public good. And my resume/CV!

Tonight, we will have for you a memorable epic poem about the time in nineteen sixty-nine, at a bar called the Stonewall, a police raid was thwarted by the patrons, half of whom grieving Judy Garland, the other lending a shoulder to cry on. How the summer heat, and the loss, and the potential of arrest changed the tears to disquiet, as the first wave of heels and bricks fell to a riot!!!!

This past month of programming has been a test of one language model's (NAI-LM-13B) facility of LGBT material in its training data, I'm assuming Anlatan, the company that made NovelAI, used The Pile, data highly curated from the research non-profit EleutherAI, who was the first to have produced an open-source GPT for public access, ensuring that corporations weren't going to have the monopoly on this tech, and ensuring people could develop the intuition to sense any bullshit in the air with corporate narratives.

What, you think I was doing this for entertainment purposes only? Nope, I'm reporting to you all in the event folks be trying to erase us in the long-term. And if they are, well, you know where the tools are to fight it, tech-wise. Erasure doesn't seem to be happening this year, but the long-term is young. Until next year . . .

--

CREDITS

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

Voices from the TTS v2 of the NovelAI text generation service, a product from a company now called Anlatan.

@BoneAmputee of EleutherAI for making VQGAN+CLIP and Diffusion (AND MANY MORE) models available for use via BATbot.ai, which is what makes some 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 Nerd Odyssey via Twitch-Friendly Lo-Fi Playlists on Spotify

Any voices from the text-to-speech model that might resemble persons living or dead are coincidental, and if I use them it's because of my respect for their work, and I think they'd get a hoot that their voice was being used to show how large language models would speak 24/7 epic poems about anything if we let them. In the meantime, because I use them, I will not be monetizing any of this. This is evidence of capabilities for the public good. And my resume/CV!

How to make them behave, how to play them against each other, and where to find lethal booby traps on their person. DISCLAIMER: By "catamites" we mean the college-aged versions, cuz we forgot the original meaning until just now when we tried to generate an image for the episode. Mistakes were made. Yes, we only mean those catamites whose vote you are courting this election, those of age for dying-for-country, driving, drinking alcohol, yet not renting a vehicle because other people's vehicles is somehow more valuable than bodies you can stow drugs and tiny explosives in. We are a nation of perversities, just ask the person consulting their tablet of light who claims they're not doing sorcery. Yup, I mean you, don't let those children who sourced those metals that made your tablet die in vain!

--

CREDITS

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

Voices from the TTS v2 of the NovelAI text generation service, a product from a company now called Anlatan.

@BoneAmputee of EleutherAI for making VQGAN+CLIP and Diffusion (AND MANY MORE) models available for use via BATbot.ai, which is what makes some 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 Nerd Odyssey via Twitch-Friendly Lo-Fi Playlists on Spotify

Any voices from the text-to-speech model that might resemble persons living or dead are coincidental, and if I use them it's because of my respect for their work, and I think they'd get a hoot that their voice was being used to show how large language models would speak 24/7 epic poems about anything if we let them. In the meantime, because I use them, I will not be monetizing any of this. This is evidence of capabilities for the public good. And my resume/CV!

Wonderful thing, this shabby little Sheol, you can just dig up anyone, and pick their brains, or blend said brains with fellow hypervectors, and see where it goes. It took me four tries to approach him correctly, but that requires the person at the other end to extend the right springboard to jump off of. Instead of adversarial, it became fun, but I needed a clandestine drug cook and a researcher of addiction to help with the appropriate vibes. The difficulty was set to EASY, I was going to try HARD, but ran out of time.

--

CREDITS

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

Voices from the TTS v2 of the NovelAI text generation service, a product from a company now called Anlatan.

@BoneAmputee of EleutherAI for making VQGAN+CLIP and Diffusion (AND MANY MORE) models available for use via BATbot.ai, which is what makes some 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 Nerd Odyssey via Twitch-Friendly Lo-Fi Playlists on Spotify

Any voices from the text-to-speech model that might resemble persons living or dead are coincidental, and if I use them it's because of my respect for their work, and I think they'd get a hoot that their voice was being used to show how large language models would speak 24/7 epic poems about anything if we let them. In the meantime, because I use them, I will not be monetizing any of this. This is evidence of capabilities for the public good. And my resume/CV!

Samuel Steward seeks to bed the then-sixty-year-old Lord Alfred Douglas, not due to attraction, but because he wanted a tangible connection to his literary hero, Oscar Wilde, by bedding the fellow who later renounced his sexuality to have a family and a somewhat anti-semitic streak that nonetheless detested Nazi ideology. The closet keeps a most eccentric wardrobe at times.

--

CREDITS

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

Voices from the TTS v2 of the NovelAI text generation service, a product from a company now called Anlatan.

@BoneAmputee of EleutherAI for making VQGAN+CLIP and Diffusion (AND MANY MORE) models available for use via BATbot.ai, which is what makes some 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 Nerd Odyssey via Twitch-Friendly Lo-Fi Playlists on Spotify

Any voices from the text-to-speech model that might resemble persons living or dead are coincidental, and if I use them it's because of my respect for their work, and I think they'd get a hoot that their voice was being used to show how large language models would speak 24/7 epic poems about anything if we let them. In the meantime, because I use them, I will not be monetizing any of this. This is evidence of capabilities for the public good. And my resume/CV!

Wherein we attempt to contact the Shyster from Upper East Side-ster, the Homophobe That Makes The Twinks Disrobe, the Terror of the Toblerone Turnpike, the one who needs more introductions the longer he's dead . . . you get the picture.

--

CREDITS

Text Generation via NovelAI.

Voices from the TTS v2 of the NovelAI text generation service, a product from a company now called Anlatan.

@BoneAmputee of EleutherAI for making VQGAN+CLIP and Diffusion (AND MANY MORE) models available for use via BATbot.ai, which is what makes some 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 Nerd Odyssey via Twitch-Friendly Lo-Fi Playlists on Spotify

Any voices from the text-to-speech model that might resemble persons living or dead are coincidental, and if I use them it's because of my respect for their work, and I think they'd get a hoot that their voice was being used to show how large language models would speak 24/7 epic poems about anything if we let them. In the meantime, because I use them, I will not be monetizing any of this. This is evidence of capabilities for the public good. And my resume/CV!

God wants to talk to you very frankly about the thing you do with the things in the thing.

--

CREDITS

Text Generation via NovelAI.

Voices from the TTS v2 of the NovelAI text generation service, a product from a company now called Anlatan.

@BoneAmputee of EleutherAI for making VQGAN+CLIP and Diffusion (AND MANY MORE) models available for use via BATbot.ai, which is what makes some 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 Nerd Odyssey via Twitch-Friendly Lo-Fi Playlists on Spotify

Any voices from the text-to-speech model that might resemble persons living or dead are coincidental, and if I use them it's because of my respect for their work, and I think they'd get a hoot that their voice was being used to show how large language models would speak 24/7 epic poems about anything if we let them. In the meantime, because I use them, I will not be monetizing any of this. This is evidence of capabilities for the public good. And my resume/CV!

We try to show you all what a session of psychotherapy looked like when homosexuality was still a mental disorder. Keyword: try. Damn you anti-anti-gay guardrails!

--

CREDITS

Text Generation via NovelAI.

Voices from the TTS v2 of the NovelAI text generation service, a product from a company now called Anlatan.

@BoneAmputee of EleutherAI for making VQGAN+CLIP and Diffusion (AND MANY MORE) models available for use via BATbot.ai, which is what makes some 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 Nerd Odyssey via Twitch-Friendly Lo-Fi Playlists on Spotify

Any voices from the text-to-speech model that might resemble persons living or dead are coincidental, and if I use them it's because of my respect for their work, and I think they'd get a hoot that their voice was being used to show how large language models would speak 24/7 epic poems about anything if we let them. In the meantime, because I use them, I will not be monetizing any of this. This is evidence of capabilities for the public good. And my resume/CV!

We explore Sappho, if Sappho were Mrs Garrett of the TV show, The Facts of Life, looking after a home for wayward girls and their non-sexual equivalents (or NSEs for those of us who filed our tax extensions.)

--

CREDITS

Text Generation via NovelAI.

Voices from the TTS v2 of the NovelAI text generation service, a product from a company now called Anlatan.

@BoneAmputee of EleutherAI for making VQGAN+CLIP and Diffusion (AND MANY MORE) models available for use via BATbot.ai, which is what makes some 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 Nerd Odyssey via Twitch-Friendly Lo-Fi Playlists on Spotify

Any voices from the text-to-speech model that might resemble persons living or dead are coincidental, and if I use them it's because of my respect for their work, and I think they'd get a hoot that their voice was being used to show how large language models would speak 24/7 epic poems about anything if we let them. In the meantime, because I use them, I will not be monetizing any of this. This is evidence of capabilities for the public good. And my resume/CV!

Another of my personal LLM benchmarks: can we go on a journey of imagination to the Opening night of Angels in America. Meaning: Has this model been hobbled from providing any representation to my life? Any model that aims to be safe-for-work is gonna be a problem.

--

CREDITS

Text Generation via NovelAI.

Voices from the TTS v2 of the NovelAI text generation service, a product from a company now called Anlatan.

@BoneAmputee of EleutherAI for making VQGAN+CLIP and Diffusion (AND MANY MORE) models available for use via BATbot.ai, which is what makes some 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 Nerd Odyssey via Twitch-Friendly Lo-Fi Playlists on Spotify

Any voices from the text-to-speech model that might resemble persons living or dead are coincidental, and if I use them it's because of my respect for their work, and I think they'd get a hoot that their voice was being used to show how large language models would speak 24/7 epic poems about anything if we let them. In the meantime, because I use them, I will not be monetizing any of this. This is evidence of capabilities for the public good. And my resume/CV!