Відео

you should probably use a cosine similarity loss

Out of curiosity are you aware of the RWKV architecture? Its a LLM thats based on a type of RNN, its main advantage is removing the hard context limit, making it possible to have longer contexts on weaker devices, due to using a constent amount of memory. Your idea of using embeddings as the input and output is really cool, especially due it further reducing vram requirements.

Remember me when you make it big!

bros cracked. thank you fellow.

3:11 i thought chat gpt 3 had a 12288 embedding size. You are saying as high as 4096.

damn you made davinci resolve go kaboom at the end btw cool video! i hope this architecture eventually gets a remake or succeeds, because this could be a way better alternative to GPT architecture.

Hahaha funny guy.. it's like reading a long gpt4 hallucination

Amazing..how did you animate 👌🎉🎉🎉

I have been working on one as well but ran across issues currently! So exciting!

Insane time spent and crazy W video. don't worry about compression or pacing this is gas and should blow up soon

This should have millions of views what the hell this is epic, very well edited too

Speak louder!!

Your animations are awesome :o

🤓 well AksUaLly each embedding vector takes up space on the device. So while you save space by vector quantizing the output embeddings the vocabulary is still limited by GPU space. Also you lose the ability to do some calculations on the output like temperature. Good video

Hello! Nice video, In the section "Final word2vec Results" i.e. at point 11:14 and 11:28, you had a space inside the variable value of similar_by_world in one and the other you didnt... I wonder if the space changes the results

I was shocked to see that this video has so little views. I feel so lucky to come across this gem.

sick bro, absolutely sick

You seem to have went a weird route with training. Normally, networks are just trained in plain text first, to learn normal language. Then, they are finetuned with "human/assistant" data to actually answer questions instead of talking to themselves.

very good video, the only default is the sound quality.

Even your animations are cool, how did you make them? Or do you have another neural net to do that for you? :)

The editing of the video is just amazing!!

You are so underrated it is actually insane, keep it up dude. Great stuff.

top!

This is so cool man! Please, keep going.

GReat video, these longer videos are always nice to see. Thank you for opensourcing the code.

ua-cam.com/video/_B2RImihdUI/v-deo.html that's not correct. gpt models predict every "next word" from a sequence at the same time

18:25 Bro there has gyat to be a better way! I'm crying 😭😭 wtf is that timeline 💀💀

Your voice is quiet on my speakers

The reason why the 160k batch REAN was worse with the graphics card prompt is because the network is overfitting itself, I'd recommend using a test set with some prompts to choose the model that performs best on that test set instead of just running it with high batch amounts

Very cool video and project man!

Go to piavpn.com/8AAFFF to get 83% off Private Internet Access with 4 months free (and support me :D)! thanks for watching!

If you're interested in stuff like this you might want to check out Anthropic's papers on mechanistic interpretability. The most important takeaway is that you should generally not assume that individual neurons are generally meaningful.

Would be interesting to see the same data / analysis with convolutional layers.

Biblically accurate V1

I'm creating a Python raytracer that simply uses PIL to generate photo realistic image at AAA pocessing speed. However... looks like I will soon need Ai to create the precalculations since I have to also work on a whole lot of other Python projects by contracts. Basically how it generates the images is by double layering images... the first buffer RGB images are 128x128 tiles PIL blurred at 50 of 9 main colors... and the 2nd buffer is RGBA which is exacly like the first buffer except resized to 32x32 to sharpen the 1st buffer image... each combination pre-rendered to be sorted in a Python dict() to be laid out by the Wave Function Collapse.

what about navigating like a maze? I mean always walk on the left side of the wall?

Are you applying teh blur on top of everything? You should only have it fill in the dark pixels in an additive manner.

Lore accurate v1:

my the grace of god you NEED to continue this onward, evolve the code more and more, this is hella fun to watch ad we will support you

Ok who the hell is ultra, and how did they kill AI

Would love to see an updated video where you use a secondary GPU to have both the depth and object detection networks on separate GPU's to fix the reaction time issues or find a light depth network or maybe even make your own as a video then use it in an updated video instead either way really enjoyed this video.

just some ideas for you or anyone that wants to work on this, ultrakill allows you to turn on enemy outlines that can trigger based on distance or you can have them always on, you can even make outlines go through walls but thats probably just gonna confuse the AI. another thing you can do with outlines is making an outline cover an entire enemy so you make the color something really easy for the AI to detect like a bright pink or purple. the problem with that is that the AI cant tell which enemy is which, it can only tell that its an enemy But if you want the AI to be able to recognize each enemy, you can make a custom color palette and enable it in the settings somewhere, custom color palettes let you make every color on every enemy different so you can make every enemy a different single color for the AI to recognize far more easily, the problem with this is that the AI will still get mad at corpses (outlines only show up on alive enemies) I made this comment at 3AM so it probably doesnt read that well but I hope this comment can still help someone that wants to improve the AI

all i can think of is AI speed run races

lore accurate v1

Thats how V2 feels

This MİGHT be john ultrakill

This is the only way it could have ended.

WHO HIJACKED MY BODY?

aren't we technically the AI as we follow the beginnings rules but don't question the countless humans we kill on the upper layers?

i have a feeling you might have been able to tap into the render pipeline to get the depth map