HN Explorer

PavleMiha

↗ HN profile [ 88.9 ms ] full profile
Karma
0
Created
()
Submissions
0
  1. No one believes a True Believer (pavlemiha.substack.com)
    2 points 0
  2. Show HN: MLGarden, a tool/toy to build and train simple neural networks visually (github.com)

    I made this tool to get some better intuition on how neural networks/backpropagation worked, but I'm really unsure what to do with it now, so I've put it up on github, and I wrote a little primer on backprop and neural…

    11 points 0
  3. Large Language Model Blues – The AI revolution made me feel bad, but I’m OK now (pavlemiha.substack.com)
    3 points 0
  4. Advice for Myself: Write More (pavlemiha.substack.com)
    3 points 1
  5. The curious case of Pretty Good human inner/outer alignment (lesswrong.com)
    9 points 0
  6. Ask HN: How much did Google's new PaLM model cost to train?

    Struggling to even think of a way to estimate this, but I'd be curious to know an order-of-magnitude estimate of how high the training costs alone were.

    1 points 0
  7. Show HN: A shop that sells 3d printed models of 67P Churyumov-Gerasminko (rosettacometmodel.com)
    2 points 1