🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

the internet is so full of robot teleop videos or 2 second clips of an "autonomous policy" (that only works on that one setup) that it's hard to put out a video of an *actually* working robot to impress normies to them a 5 minute video of a robot actually working is just boring

numerology but for training performance based on SLURM job IDs

ready for another beautiful day of semi-automated ML engineering it's crazy how rapidly things have changed

really wish Weights & Biases had actual job state callbacks instead of forcing me to write polling infrastructure

we need Chinchilla-style fixed compute budget trade-off curves across - model size - pre-training compute - RL compute I suspect they'd show Grok4 spent too much time doing RL, not enough time pre-training. No idea what to expect on model size.

get a job that lets you do both