NN-Z2H: Lecture 3: Building makemore Part 2: MLP

We implement a multilayer perceptron (MLP) character-level language model. In this video we also introduce many basics of machine learning (e.g. model training, learning rate tuning, hyperparameters, evaluation, train/dev/test splits, under/overfitting, etc.). Read more ›

More Articles

• NN-Z2H: Lecture 2: The spelled-out intro to language modeling: building makemore
• NN-Z2H: Lecture 1: The spelled-out intro to neural networks and backpropagation: building micrograd
• Structured Pruning Learns Compact and Accurate Models
• Augmenting Language Models with Long-Term Memory
• How to improve graduate application after submission
• AlpaServe: Statistical Multiplexing with Model Parallelism for Deep Learning Serving
• Read List of LLM
• High-throughput Generative Inference of Large Language Models with a Single GPU
• Report of the 2023 UIUC SE Evaluation Task
• LiteFlow: Towards High-performance Adaptive Neural Networks for Kernel Datapath
• Tips for Success as a New Researcher
• Clio: A Hardware-Sofuware Co-Designed Disaggregated Memory System
• SIGOPS-2005: The Structure of the "THE"-Multiprogramming
• A Medium-Grained Algorithm for Distributed Sparse Tensor Factorization
• SIGOPS: The Hall of Fame Award
• Show All Articles ›