Mar 21 · 5 min read · I've been running autonomous agents on transformer-based models for months. The inference bill is brutal. Every self-improvement cycle our RSI engine runs - measure, hypothesize, mutate, test, apply or discard - costs real money in GPU time. And most...
Join discussionFeb 3 · 5 min read · Recently Mamba has been making waves due to it’s linear time complexity in regards to processing tokens sequential. It is basically a Linear RNN under the hood but with selective forgetting and selective memorization, the very ablity that sets the Tr...
Join discussion
Mar 5, 2024 · 4 min read · !UPDATE: i'm now publishing here Introduction In Python data science and machine learning, package and environment management systems are indispensable. Conda, initially birthed by Anaconda, is an open-source manager for packages and environments. It...
Join discussionAug 7, 2023 · 3 min read · Summary: There's nothing more fun than setting up Python environments and packages. I started using Micromamba after getting angry at a previous mamba/conda/WSL installation (self-induced pain/complication). I wanted to try it in hopes of finding a s...
Join discussion
