NanoEuler
GPT-2 scale model in pure C/CUDA from scratch. GPT-2-style LLM built from scratch in C/CUDA with hand-written backprop, BPE tokenizer, FlashAttention, pretraining, and SFT. The two interesting reasons that led me to create nanoeuler were (1) interfacing with llm does not mean understanding how they are composed and (2), working on llm with a very low-level layer to understand the correlation between parameters and data and growth of the model and how the GPU works and.
오픈소스
반응 (2)
비슷한 프로덕트
설명이 비슷한 순
UNet diffusion model in pure CUDA
github.comUNet diffusion model in pure CUDA
Simple UI for LLM Model Finetuning
Tiny-vLLM
github.comhigh performance LLM inference engine in C++ and CUDA
DPO, GRPO)
WebGPU FTW with Splat-Transform
github.comCLI tool and library for 3D Gaussian splat processing and conversion