-
Retrieval-Augmented Generation: Chat With Your Documents
Build a Retrieval-Augmented Generation (RAG) pipeline that lets the model answer questions from your own documents.
-
Building a Chat Harness: From Model to Product
Build a chat harness with conversation history, system prompts, and chat templates — the missing piece between a…
-
Real Models: Running GPT-2 Locally
Run real pre-trained models (GPT-2) locally using HuggingFace. Understand the difference between a model and a product.
-
Text Generation: From Logits to Words
Explore text generation strategies: greedy decoding, temperature sampling, top-k, and top-p (nucleus) sampling.
-
Training a Transformer: Making It Learn
Train your transformer using PyTorch. Watch the loss drop from 3.5 to 0.13 in 30 seconds on CPU.
-
Building a Transformer From Scratch
Assemble embeddings, attention, and feed-forward layers into a complete decoder-only transformer (like GPT).
-
Self-Attention: The Engine Behind Transformers
Learn how self-attention works — the key mechanism that lets every token in a sequence look at every…
-
Embeddings: Turning Words Into Numbers
Understand how tokens become dense vectors that capture meaning, and why one-hot encoding is the wrong approach.
Categories
Tag Cloud
.net algorithms angular api Array arrays basic-concepts big o blazor c# classes code components containers control-structures csharp data structures data types Deep Learning dictionaries docker dom dotnet framework functions git guide javascript json leetcode linq lists LLM methods MVC Natural Language Processing npm object oriented programming oop operators Python sorted tutorial typescript web framework