2 items tagged “vicki-boykis”
2024
GGUF, the long way around (via) Vicki Boykis dives deep into the GGUF format used by llama.cpp, after starting with a detailed description of how PyTorch models work and how they are traditionally persisted using Python pickle.
Pickle lead to safetensors, a format that avoided the security problems with downloading and running untrusted pickle files.
Llama.cpp introduced GGML, which popularized 16-bit (as opposed to 32-bit) quantization and bundled metadata and tensor data in a single file.
GGUF fixed some design flaws in GGML and is the default format used by Llama.cpp today.
2021
Writing for distributed teams (via) Vicki Boykis describes how she only sent 11 emails during her first 12 months working at Automattic, because the company culture there revolves around asynchronous communication through durable writing using the P2 custom WordPress theme. “This is a completely different paradigm than I’ve ever worked in, which has been a world usually riddled with information lost to Slack, Confluence, and dozens of email re:re:res.”