Это видео недоступно.
Сожалеем об этом.
Exploring Mini GPT-4: Multimodal LLM with Open Source Tools
HTML-код
- Опубликовано: 24 апр 2023
- In this video, we dive into MiniGPT-4, a powerful application that combines open source tools to describe images in text. We explore its model architecture, training process, and the fascinating concept of soft prompts. Discover how this application pushes the boundaries of large language models and their multimodal capabilities.
🔗 MiniGPT-4 Paper: arxiv.org/pdf/...
🔗 MiniGPT-4 Project Page: minigpt-4.gith...
🔗 Repository: github.com/Vis...
About me:
Follow me on LinkedIn: / csalexiuk
Check out what I'm working on: getox.ai/
#minigpt4 #gpt4 #multimodal #llm
Excelent video I hope you can make a similar one about Multimodal-GPT which I think is based on this one and other multimodal projects. Love your content! Thanks so much for your contribution.
I'll definitely take a look at that!
Hey Chirs. I just read the paper, and I have a few questions If you don't mind. Do you know if there is any way to access the weights of the progression layer? I'm doing a small project where I change the frozen vicuma model with a different model, but I don't know how to extract the weighys of the progression layers. If you have any insight, it would be awesome. PS: I tried to contact the project developers, but so far no sucess.
Misleading thumbnail. I spent 15 minutes waiting for the explanation to end so that I could get some laughs at AIs trying to explain memes.........then the video was over.......
Totally fair!
The opening was the AI trying to explain the Spiderman meme, but I definitely could've spent more time having it explain the memes, sorry Sven!
I've made adjustments to the thumbnail, hopefully this time it's less misleading - thanks for the feedback.
@@chrisalexiuk np, the content is good but not what I was expecting. I will say it could still have that visual appeal of the meme even with the new text. Just change the arrow text to "why is this funny?" or something.
Awesome feedback, will implement!
Great stuff Chris
Thanks, Panos!