p-e-w /
cliCLI tool that removes safety alignment from transformer language models automatically, using directional ablation without post-training.
01 / THE LONG VIEW
Star counts recorded on every board appearance.
02 / DAY BY DAY
Explore the calendar for changes between observed star counts.
03 / THE SOURCE RECORD
Where it appeared, the rank it held, and the board that noticed.
Keep exploring / Related repositories
More from the collection, connected by their recorded descriptions, topics, or language.
MakazhanAlpamys /
A command-line tool that fine-tunes and post-trains LLMs locally from a single YAML config, including low-VRAM layer streaming.
FlashML-org /
Edge-native MoE serving engine that runs frontier-scale open-weight models on consumer GPUs and CPUs.
JustVugg /
Pure-C inference engine that runs very large mixture-of-experts models across VRAM, RAM, and storage tiers on consumer hardware.