LoreonLabsPlatform

Overview

Intelligence

Markets
Builders
Research
Narratives
Ecosystems
Launchpads

Discover

Search
Sources

Python

nanochat

The best ChatGPT that $100 can buy.

PythonEmerging

Stars

—

Forks

—

Contributors

8

Last push

6d ago

Recent commits

Latest commits.

Add comprehensive MoE documentation with benchmarks and scaling analysis
53ce4f7ademeure2mo ago
Add --num-experts, --top-k, --num-shared-experts CLI args to base_train.py
c0e579eademeure2mo ago
Fix MoE dtype compatibility with COMPUTE_DTYPE system
674d623ademeure2mo ago
Add FA3/FlexAttention/SDPA attention backend (cherry-picked from fa3-flex-sdpa branch)
e006f7fademeure2mo ago
Merge upstream/master into moe-new: add smear/backout, GradScaler, ClimbMix, autoresearch tuning
ccb6dbdademeure2mo ago

create a group for dev dependencies, there is no need to install all this other stuff just for speedrun and it's exposing people to dependency chain attacks. we need to delete more dependencies. dependencies bad bad bad

a445144Andrej Karpathy3mo ago

delete non-essential deps from legacy use

03be953Andrej Karpathy3mo ago

Merge pull request #595 from svlandeg/fix/typo

7808dc7Andrej3mo ago

Top contributors

Builders behind this project.

ericsilberstein