amrb

25 ms·

1. 241.▲
   by amrb 4y ago

   > Any suggestion or idea or process how to start this kind of VC fund? Your gonna need to convince someone somewhere that the returns from the investment, are gonna be better than the interest rate of today, maybe that won't be a ha
2. 242.▲
   by amrb 4y ago

   Having open source models could be as important as the Linux project imo
3. 243.▲
   by amrb 4y ago

   Thanks for put this out there! Open question, does the written by ML model thing feel like catching people wearing a fake rolex, I can argue it two ways here. So we have tools like google, calculators and now GPT.. the big but here being if
4. 244.▲
   by amrb 4y ago

   High street means Main street for out American viewers :)
5. 245.▲
   by amrb 4y ago

   Should we be more worried that phone don't really have a kids profile or that mobile games < PC and that kid should be playing CS and StarCraft!
6. 246.▲
   by amrb 4y ago

   I really hope this does not become a form of nation state trolling, letting off hundreds of balloons to mess with the worlds airspace.
7. 247.▲
   by amrb 4y ago

   Bro you see the part where I talk about swapping out openai. You can download ML models from the site hugging face.co, if you read this link you should be well on your way to building something today https://github.com/huggi
8. 248.▲
   by amrb 4y ago

   Here's a guide to setup a server the like hundreds of people already have, like why post articles about such obviously things https://marketplace.digitalocean.com/apps/mastodon
9. 249.▲
   by amrb 4y ago

   The secret ingredient is human feedback, would you invest billions in 40 people in a call center clicking between two text responses, as between that and scraped data from the internet it's not magic. To answer the question it's t
10. 250.▲
    by amrb 4y ago

    The idea of a personal database is great, it could be books you read, articles or in this case advice but I find with memory being unreliable having a quick was to search is great!
11. 251.▲
    by amrb 4y ago

    There's also the option of caching and sessions with https://github.com/HazyResearch/manifest
12. 252.▲
    by amrb 4y ago

    Was it not the joke, that Stanford people need everyone to know it
13. 253.▲
    by amrb 4y ago

    Talk about bad omens!
14. 254.▲
    by amrb 4y ago

    You can swap openai for models on huggingface.co but the quality is not equal yet.
15. 255.▲
    by amrb 4y ago

    You are 100% right, people are building on sand if they think any VC will fund any who doesn't own the model underpinning there products.
16. 256.▲
    by amrb 4y ago

    "Do you want megacorps? Because that's how you get megacorps" -Archer Control is why I want to see more "open source" version of machine learning models be it over data or auditing the output and if I can p2p downlo
17. 257.▲
    by amrb 4y ago

    I've already built this and you can use this as a guide, being PII we just need to swap openai for a model on huggingface.co, this let's you do the question answering part run locally. https://dagster.io/blog&#x2F
18. 258.▲
    by amrb 4y ago

    I'm going to coin the cyber-security finding "Bill exhaustion via open ML prompts".. wonder how long till it's in the Owasp top X
19. 259.▲
    by amrb 4y ago

    Posted Jan 9th insert slowpoke meme https://dagster.io/blog/chatgpt-langchain
20. 260.▲
    by amrb 4y ago

    A weapon to surpass metal gear!!
21. 261.▲
    by amrb 4y ago

    Agreed its can lacking logic till to spell it out, example was this ctf challenge and it just could understand the hash collision till I gave it the full write up [0] w.r.t codebases I may look at some of the free models (as this gets aroun
22. 262.▲
    by amrb 4y ago

    Silly idea tho would it be helpful in creating the server side code for an online only game? I've seen a project for battlefield 3 tho already have the feeling it's a team effort at minimum?
23. 263.▲
    by amrb 4y ago

    From my own research, ctf tools like angr can build AST trees, so I'm working on the thinking I can train the AI to review interesting parts of the execution tree. happy to get feedback or papers since this has been the most interestin
24. 264.▲
    by amrb 4y ago

    Assume it's a binary blob that's needed, given it mentions Vulkan so they are skipping the ROCm method too.
25. 265.▲
    by amrb 4y ago

    Langchain has been a good project in the area: https://github.com/hwchase17/langchain
26. 266.▲
    by amrb 4y ago

    I recommend this video on the subject. "Let's build GPT: from scratch, in code, spelled out." https://youtu.be/kCc8FmEb1nY
27. 267.▲
    by amrb 4y ago

    On the Internet, nobody knows you're a dog!
28. 268.▲
    by amrb 4y ago

    I've been working on a hobby project to make all the site, videos and conversations I've seen searchable (aka external memory). While building it I used this to prepare the data and get around the prompt limits https://
29. 269.▲
    by amrb 4y ago

    Could be a good weekend project, they already allow importing keys for decryptions so most of the work is finding references where openssl, etc stores it.
30. 270.▲
    by amrb 4y ago

    Sounds like a good weekend project, investigate the effects of disabling ebpf.