I packaged up the "autoresearch" project into a new self-contained minimal repo if people would like to play over the weekend. It's basicall

kept by eddie mar 08

A self-contained repo allows AI agents to autonomously iterate on LLM training code, aiming for indefinite progress in lowering validation loss without human involvement.

Three days ago I left autoresearch tuning nanochat for ~2 days on depth=12 model. It found ~20 changes that improved the validation loss. I
An autonomous research agent discovered ~20 additive training improvements for nanochat, cutting the leaderboard's 'Time to GPT-2' by 11% from 2.02 to 1.80 hours.
also on nanochat, #machine-learning, #automation, #open-source
Autoresearch on an old research idea | Blog
Using Karpathy's Autoresearch loop with Claude Code, the author reduced mean rank by 54% on an old research project, primarily through bug fixes and hyperparameter tuning.
also on autoresearch, #machine-learning, #ai-agents, #open-source
'The Karpathy Loop': 700 experiments, 2 days, and a glimpse of where AI is heading
Andrej Karpathy's 'autoresearch' AI agent ran 700 experiments in 2 days, discovering 20 optimizations that sped up language model training by 11%.
also on autoresearch, #automation, #ai-agents