LLM - Afternoon with Green Tea

Sign in

LLM

large language model

Karpathy's McroGPT

Recently, Andrej Karpathy released a tiny file that quietly shook a large part of the AI community: a complete GPT implementation written in 243 lines of pure Python. No PyTorch. No TensorFlow. No NumPy acceleration. Just math. The project — often called microGPT — demonstrates that a modern language model is not

Model training and optimization

Understanding GPU Memory Anatomy (Hugging Face Guide), referring to the reference: https://huggingface.co/docs/transformers/model_memory_anatomy

Code generation via LLM

From Text Completion to Agentic Software Engineering

SIGIR-Style Background: Retrieval-Augmented & Augmented Language Models

Paper summary from Sigir 2023

DSPy Analysis

DSPy by Stanford: Programming (and Compiling) LM Pipelines into Self-Improving Systems

The Evolution of Mamba: From Selective State Spaces to Linear-Time Reasoning Models

Llama 2 and Llama 3 Analysis

Meta has released Llama 2 (model and finetune tutorial) recently, and one of my research works is possible to leverage its text completion capability for synthetic data generation.Llama 2: Open Foundation and Fine-Tuned Chat Models, I'd like to draft this blog to understand more details under the

Augmented Language Models: a Survey

A important survey for ALM from Yann LeCunn