Stories from the gateway
Fundamentals
Positional Encoding: How a Transformer Knows Token Order
Attention is order-blind by default. How position gets injected, why the method decides how far context can stretch, and what you see when it fails.
· 9 min read
Fundamentals
Pretraining vs Post-Training: Where a Model Gets Its Behaviour
Pretraining decides what a model knows. Post-training decides how it behaves. Knowing which stage owns a problem tells you whether prompting can fix it.
· 9 min read
Fundamentals
Repetition Penalties: Why They Break Code Generation
· 9 min read
Fundamentals
Residual Connections: Why Deep Models Train at All
· 8 min read
Fundamentals
RLHF Explained: Training on Comparisons Instead of Answers
· 9 min read
Fundamentals
Rotary Embeddings (RoPE) Explained With Clock Hands
· 9 min readFundamentals
Scaling Laws Explained: What Predicts a Model Getting Better
· 9 min read
Fundamentals
Self-Consistency Decoding: Ask Five Times, Take the Majority
· 8 min readFundamentals
Sparse vs Dense Activation: What Runs on Every Token
· 9 min readFundamentals
Speculative Decoding in Practice: When It Pays Off
· 9 min readFundamentals
Stop Sequences Explained: Ending Generation on Purpose
· 8 min readFundamentals
Synthetic Training Data: What It Does and Where It Fails
· 9 min readFundamentals
Test-Time Compute: Buying Accuracy With Tokens Instead of Size
· 9 min readFundamentals
The Lost-in-the-Middle Problem: Position Beats Relevance
· 8 min readReady to start?
Every model, one key. Start free and upgrade when you are ready.