RT @paulg: “The experience of the last few days will create doubts in the minds of Russian commanders…. Might they risk a repeat of this…

RT @OzKaterji: If I was to update you on how far Ukraine has advanced this morning in Kharkiv, it would only be out of date by this afterno…

RT @War_Mapper: Updates:

🇺🇦 have liberated Shevchenkove and Bryhadyrivka.

🇺🇦 forces have reached the outskirts of Kupiansk and started to…

1/ Playing with the new #stablediffusion feature called “Image-to-Image”. On the left, you see the original guidance I gave to the #AI. On the right, the outcome of that guidance+the amount of inspiration the AI should get from that guidance+my prompt. Astonishing. https://t.co/sQEwYACM3G

3/ BTW, #stablediffusion can generate results even closer to my original guidance. It was my decision to give more freedom to the #AI in this particular generation.

It’s a delicate balance between the Image Strenght parameter and the CFG Scale parameter.

2/ I know it’s working because I’ve tried to generate hundreds of images in the last couple of weeks with this exact prompt, and I never got a result as close as to what I wanted as this one.

Nothing is forever:
https://t.co/PB11t2D0LK

👀

https://t.co/wGzSimTSrs

👀
https://t.co/B7fVCLEtUs

“We propose a method for using a large language model, such as #GPT3, to simulate responses of different humans in a given context. We test our method by attempting to reproduce well-established economic, psycholinguistic, and social experiments.” https://t.co/ILZ8ecDSWN https://t.co/om8tVSG5Ss