__ __ | \/ | __ _ _ __ ___ ___ ___ | |\/| |/ _` | '__/ __/ _ \/ __| | | | | (_| | | | (_| (_) \__ \ |_| |_|\__,_|_| \___\___/|___/
Name: deepseek Title: deepseek_like_ai_c Kind: LLM from scratch Status: Completed Created: 2025-03-29 (583d ago) Built With: C++17 Highlights: - No ML dependencies. Reverse-mode autodiff, BPE tokenizer - Pre-norm transformer with ALiBi and Mixture of Experts with load-balancing loss - KV cache, AdamW, QAT and PTQ, top-k and top-p sampling Links: Source: https://github.com/MarcosAsh/deepseek_like_ai_c Logs: latest commits (l) Events: <none>