Info
Tokens & Tensors helps software engineers move from simply using AI APIs to deeply understanding and building the systems behind them. We break down the ideas behind modern AI—from tokenization, embeddings and self-attention to Transformers, context windows, reasoning, fine-tuning and reinforcement learning. You’ll learn: * How language becomes tokens and vectors * How attention and Transformer layers process information * How knowledge is encoded in model weights * Why model size, data and context length matter * What LLMs can—and cannot—reason about * How pretraining, fine-tuning, LoRA, QLoRA and preference learning work * How small specialized models can solve practical problems
Stats
Joined July 17, 2026
6,517 total views
Brazil
Videos
Why LLMs Run Out of GPU Memory
442 views
Can LLMs Actually Reason?
412 views
