正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

added configurable color ink accents + thickness 🧑‍🍳 again, the difference with this is that the output timing matches 1:1 with your input (at 0.5x below) still jus' css + svg 💫

jhey ʕ•ᴥ•ʔ

146,437 subscribers

32,524 次观看 • 3 个月前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

That SVG filter is the whole reason this CSS scroll animation is even possible 🙏 (now with improved timing ✨)

That SVG filter is the whole reason this CSS scroll animation is even possible 🙏 (now with improved timing ✨)

jhey ʕ•ᴥ•ʔ

94,093 次观看 • 2 年前

you can totally do it with css 🧑‍🍳 use a border mask, layers + overflow: hidden, or other tricks skip animating the gradient though, jus' spin a 1:1 gradient layer 🤙

you can totally do it with css 🧑‍🍳 use a border mask, layers + overflow: hidden, or other tricks skip animating the gradient though, jus' spin a 1:1 gradient layer 🤙

jhey ʕ•ᴥ•ʔ

286,399 次观看 • 9 个月前

an alternative approach to console animation that doesn't hijack the console 1. inline CSS animation into an SVG 2. style console message with background-image 3. use SVG as the background-image 🚀 console.info( "%ccome cook", "background-image: url(/animated.svg);" )

an alternative approach to console animation that doesn't hijack the console 1. inline CSS animation into an SVG 2. style console message with background-image 3. use SVG as the background-image 🚀 console.info( "%ccome cook", "background-image: url(/animated.svg);" )

jhey ʕ•ᴥ•ʔ

93,420 次观看 • 1 年前

this could've jus' been a checkbox, but they added a slider with a little friction at either end that reflects the physical behavior tiny details, thoughtfully considered

this could've jus' been a checkbox, but they added a slider with a little friction at either end that reflects the physical behavior tiny details, thoughtfully considered

jhey ʕ•ᴥ•ʔ

94,368 次观看 • 9 天前

pov you’re spending the night with that one furry you’ve been playing at your local round 1 with

pov you’re spending the night with that one furry you’ve been playing at your local round 1 with

Takuma

23,026 次观看 • 8 个月前

8/1 Mr Pool posted this twice in August. 👀 This lines up with bearableguy123 calendar and with the Clarity Act timing. XRP is primed right now💥🚀

8/1 Mr Pool posted this twice in August. 👀 This lines up with bearableguy123 calendar and with the Clarity Act timing. XRP is primed right now💥🚀

Seer

28,223 次观看 • 1 个月前

a well-known iOS Safari quirk is that inputs with a font-size <16px trigger an auto-zoom on focus using 16px is probably the best solution, but it can feel a bit heavy in tight UIs a clever workaround is to keep the font-size at 16px and scale the input down with CSS!

a well-known iOS Safari quirk is that inputs with a font-size <16px trigger an auto-zoom on focus using 16px is probably the best solution, but it can feel a bit heavy in tight UIs a clever workaround is to keep the font-size at 16px and scale the input down with CSS!

daniel petho

150,773 次观看 • 5 个月前

A look at the 5 new completely custom weapons we've added to Runeteria!💀These will either be dropped directly or crafted with materials from our dungeon "Hollow Crypt"☠️We've added custom particles and trails along with emissive textures to give the weapons that extra skeletal finish👌 Also notice the custom signature ability added to our first UNIQUE item💫 Aumo Has been cooking🧑‍🍳 Stay tuned for the first look at our custom made boss fight with some spicy custom mechanics! #Hytale

A look at the 5 new completely custom weapons we've added to Runeteria!💀These will either be dropped directly or crafted with materials from our dungeon "Hollow Crypt"☠️We've added custom particles and trails along with emissive textures to give the weapons that extra skeletal finish👌 Also notice the custom signature ability added to our first UNIQUE item💫 Aumo Has been cooking🧑‍🍳 Stay tuned for the first look at our custom made boss fight with some spicy custom mechanics! #Hytale

Runeteria

16,160 次观看 • 7 个月前

[LSTM] by Hand ✍️ LSTMs have been the most effective architecture to process long sequences of data, until our world was taken over by the Transformers. LSTMs belong to the broader family of recurrent neural network (RNNs) that process data sequentially in a recurrent manner. Transformers, on the other hand, abandon recurrence and use self-attention instead to process data concurrently in parallel. Recently, there is renewed interest in recurrence as people realized self-attention doesn’t scale to extremely long sequences, like hundreds of thousands of tokens. Mamba is a good example to bring back recurrence. All of a sudden, it is cool to study LSTMs. How do LSTMs work? [1] Given ↳ 🟨 Input sequence X1, X2, X3 (d = 3) ↳ 🟩 Hidden state h (d = 2) ↳ 🟦 Memory C (d = 2) ↳ Weight matrices Wf, Wc, Wi, Wo Process t = 1 [2] Initialize ↳ Randomly set the previous hidden state h0 to [1, 1] and memory cells C0 to [0.3, -0.5] [3] Linear Transform ↳ Multiply the four weight matrices with the concatenation of current input (X1) and the previous hidden state (h0). ↳ The results are feature values, each is a linear combination of the current input and hidden state. [4] Non-linear Transform ↳ Apply sigmoid σ to obtain gate values (between 0 and 1). • Forget gate (f1): [-4, -6] → [0, 0] • Input gate (i1): [6, 4] → [1, 1] • Output gate (o1): [4, -5] → [1, 0] ↳ Apply tanh to obtain candidate memory values (between -1 and 1) • Candidate memory (C’1): [1, -6] → [0.8, -1] [5] Update Memory ↳ Forget (C0 .* f1): Element-wise multiply the current memory with forget gate values. ↳ Input (C’1 .* o1): Element-wise multiply the “candidate” memory with input gate values. ↳ Update the memory to C1 by adding the two terms above: C0 .* f1 + C’1 .* o1 = C1 [6] Candiate Output ↳ Apply tanh to the new memory C1 to obtain candidate output o’1. [0.8, -1] → [0.7, -0.8] [7] Update Hidden State ↳ Output (o’1 .* o1 → h1): Element-wise multiply the candidate output with the output gate. ↳ The result is updated hidden state h1 ↳ Also, it is the first output. Process t = 2 [8] Initialize ↳ Copy previous hidden state h1 and memory C1 [9] Linear Transform ↳ Repeat [3] [10] Update Memory (C2) ↳ Repeat [4] and [5] [11] Update Hidden State (h2) ↳ Repeat [6] and [7] Process t = 3 [12] Initialize ↳ Copy previous hidden state h2 and memory C2 [13] Linear Transform ↳ Repeat [3] [14] Update Memory (C3) ↳ Repeat [4] and [5] [15] Update Hidden State (h3) ↳ Repeat [6] and [7]

[LSTM] by Hand ✍️ LSTMs have been the most effective architecture to process long sequences of data, until our world was taken over by the Transformers. LSTMs belong to the broader family of recurrent neural network (RNNs) that process data sequentially in a recurrent manner. Transformers, on the other hand, abandon recurrence and use self-attention instead to process data concurrently in parallel. Recently, there is renewed interest in recurrence as people realized self-attention doesn’t scale to extremely long sequences, like hundreds of thousands of tokens. Mamba is a good example to bring back recurrence. All of a sudden, it is cool to study LSTMs. How do LSTMs work? [1] Given ↳ 🟨 Input sequence X1, X2, X3 (d = 3) ↳ 🟩 Hidden state h (d = 2) ↳ 🟦 Memory C (d = 2) ↳ Weight matrices Wf, Wc, Wi, Wo Process t = 1 [2] Initialize ↳ Randomly set the previous hidden state h0 to [1, 1] and memory cells C0 to [0.3, -0.5] [3] Linear Transform ↳ Multiply the four weight matrices with the concatenation of current input (X1) and the previous hidden state (h0). ↳ The results are feature values, each is a linear combination of the current input and hidden state. [4] Non-linear Transform ↳ Apply sigmoid σ to obtain gate values (between 0 and 1). • Forget gate (f1): [-4, -6] → [0, 0] • Input gate (i1): [6, 4] → [1, 1] • Output gate (o1): [4, -5] → [1, 0] ↳ Apply tanh to obtain candidate memory values (between -1 and 1) • Candidate memory (C’1): [1, -6] → [0.8, -1] [5] Update Memory ↳ Forget (C0 .* f1): Element-wise multiply the current memory with forget gate values. ↳ Input (C’1 .* o1): Element-wise multiply the “candidate” memory with input gate values. ↳ Update the memory to C1 by adding the two terms above: C0 .* f1 + C’1 .* o1 = C1 [6] Candiate Output ↳ Apply tanh to the new memory C1 to obtain candidate output o’1. [0.8, -1] → [0.7, -0.8] [7] Update Hidden State ↳ Output (o’1 .* o1 → h1): Element-wise multiply the candidate output with the output gate. ↳ The result is updated hidden state h1 ↳ Also, it is the first output. Process t = 2 [8] Initialize ↳ Copy previous hidden state h1 and memory C1 [9] Linear Transform ↳ Repeat [3] [10] Update Memory (C2) ↳ Repeat [4] and [5] [11] Update Hidden State (h2) ↳ Repeat [6] and [7] Process t = 3 [12] Initialize ↳ Copy previous hidden state h2 and memory C2 [13] Linear Transform ↳ Repeat [3] [14] Update Memory (C3) ↳ Repeat [4] and [5] [15] Update Hidden State (h3) ↳ Repeat [6] and [7]

Tom Yeh

73,030 次观看 • 2 年前

🆕 React Native Micro Skill release🚨: Expo Router App Store Card Transition. Get the skill in link below 👇 This is combo of Expo Router Zoom transitions combined with the right card format plus a landing native screen that matches the effect. Also cards bg text matches image with blur-hash. (Text color switch depending on that color too). Thats it. I call these "micro skills" and that's the point. This small skill artifacts is very simple in sight but still deep in its own way. It is thinking about how to make the effect powerful. What are the other elements to combine to make a a great UX show into existence. Again micro skills are so powerful when they compound and are what will push you into creating amazing native feel experiences. Are you creating micro skills too? LMK in the comments below.

🆕 React Native Micro Skill release🚨: Expo Router App Store Card Transition. Get the skill in link below 👇 This is combo of Expo Router Zoom transitions combined with the right card format plus a landing native screen that matches the effect. Also cards bg text matches image with blur-hash. (Text color switch depending on that color too). Thats it. I call these "micro skills" and that's the point. This small skill artifacts is very simple in sight but still deep in its own way. It is thinking about how to make the effect powerful. What are the other elements to combine to make a a great UX show into existence. Again micro skills are so powerful when they compound and are what will push you into creating amazing native feel experiences. Are you creating micro skills too? LMK in the comments below.

RO

15,704 次观看 • 1 个月前

Theres a difference between snitching and being complicit with criminal activity. Ill never snitch, but when someone is borderline ruining the sport I love I’m going to fight for the greater good. #GrabGate is the biggest story heading into round 1 NCAA Lacrosse John Wick NSA/CSS

Theres a difference between snitching and being complicit with criminal activity. Ill never snitch, but when someone is borderline ruining the sport I love I’m going to fight for the greater good. #GrabGate is the biggest story heading into round 1 NCAA Lacrosse John Wick NSA/CSS

Nick Ossello

54,401 次观看 • 1 年前

China is literally on 🔥 Baidu from China has launched ERNIE 4.5 and ERNIE X1 and it’s freaking cheap . Here is everything you need to know. ERNIE 4.5 - Native multimodal and Outperforms GPT 4.5 in multiple benchmarks at just 1% of GPT 4.5 price - OpenAI GPT 4.5 – Input: $75 / 1M tokens, Output: $150 / 1M tokens; - ERNIE 4.5 – Input: $0.55 / 1M tokens, Output: $2.20 / 1M tokens ERNIE X1 - A deep thinking reasoning model with multimodal capabilities on par with DeepSeek R1 at only half the price See it in action and check out the pricing details👇 📹 source : yiyan[.]baidu[.]com 1/6 ERNIE 4.5 is a multimodal which can take Audio files as well.

China is literally on 🔥 Baidu from China has launched ERNIE 4.5 and ERNIE X1 and it’s freaking cheap . Here is everything you need to know. ERNIE 4.5 - Native multimodal and Outperforms GPT 4.5 in multiple benchmarks at just 1% of GPT 4.5 price - OpenAI GPT 4.5 – Input: $75 / 1M tokens, Output: $150 / 1M tokens; - ERNIE 4.5 – Input: $0.55 / 1M tokens, Output: $2.20 / 1M tokens ERNIE X1 - A deep thinking reasoning model with multimodal capabilities on par with DeepSeek R1 at only half the price See it in action and check out the pricing details👇 📹 source : yiyan[.]baidu[.]com 1/6 ERNIE 4.5 is a multimodal which can take Audio files as well.

AshutoshShrivastava

324,324 次观看 • 1 年前

[1/8] Drawing → 3D render with Gemini 2.0 image generation… by Dev Valladares + me 👇🧵Make your own at the link below

[1/8] Drawing → 3D render with Gemini 2.0 image generation… by Dev Valladares + me 👇🧵Make your own at the link below

Trudy Painter

71,766 次观看 • 1 年前

“Better That Way” with Luke Combs is all yours this Friday, May 15th. Presave now at the link below.

“Better That Way” with Luke Combs is all yours this Friday, May 15th. Presave now at the link below.

Charles Wesley Godwin

14,064 次观看 • 4 个月前

Interviewer: "This tournament is 1 v 1... Who do you like to play 1 v 1 the most?" Luka: "At practices we play a lot. One that I played with and it was really something unbelievable was Kyrie. He is really the kinda of 1 v 1 player who don't exist in the world. So with him was most beautiful to play" Source:

Interviewer: "This tournament is 1 v 1... Who do you like to play 1 v 1 the most?" Luka: "At practices we play a lot. One that I played with and it was really something unbelievable was Kyrie. He is really the kinda of 1 v 1 player who don't exist in the world. So with him was most beautiful to play" Source:

Luka Updates

36,151 次观看 • 3 个月前

Yomif Kejelcha breaking the 2-hr barrier on his first official marathon at London 2026 with a time of 1:59:41…but still losing to Sabastian Sawe at 1:59:30 is the long-distance running equivalent of that Jerry Seinfeld Olympics bit:

Yomif Kejelcha breaking the 2-hr barrier on his first official marathon at London 2026 with a time of 1:59:41…but still losing to Sabastian Sawe at 1:59:30 is the long-distance running equivalent of that Jerry Seinfeld Olympics bit:

Trung Phan

49,820 次观看 • 4 个月前

It's still really insane to me that this is fanmade. The quality of the animation is literally on par with the actual Clone Wars show and that's super impressive. Also the color schemes of the Clones are amazing :)

It's still really insane to me that this is fanmade. The quality of the animation is literally on par with the actual Clone Wars show and that's super impressive. Also the color schemes of the Clones are amazing :)

MA5K CEO

1,011,830 次观看 • 2 年前

Entirely done using only a phone This is Move AI in action by Olly Hayes over at Lolly Studio in the UK Mocap has come a long way. I’m combination with tools like Simulon the future of driving synths with human input is looking bright. Links below ⬇️

Entirely done using only a phone This is Move AI in action by Olly Hayes over at Lolly Studio in the UK Mocap has come a long way. I’m combination with tools like Simulon the future of driving synths with human input is looking bright. Links below ⬇️

Linus ✦ Ekenstam

83,037 次观看 • 3 年前