Interactive Learning Course

LLM Fundamentals

Master how large language models work -- from neurons to transformers to RLHF. Interactive exercises, code sandboxes, and AI-powered evaluation.

22
Modules Ready
25
Total Modules
0
Concepts Learned
--
Avg Score
Overall Progress: 0%
1
Architecture
9 modules -- How transformers work from the ground up
1
Introduction: How LLMs Work
Neural networks, forward pass, training loop, next-word prediction
Ready
->
2
Tokenization
Characters to tokens, BPE algorithm, byte-level encoding
Ready
->
3
The Embedding Layer
One-hot vs dense embeddings, semantic space, king-queen arithmetic
Ready
->
4
Attention
Query/Key/Value, dot products, softmax, multi-head attention
Ready
->
5
Positional Encoding
Sinusoidal encoding, RoPE, why word order matters
Ready
->
6
Layers of Understanding
Residual connections, layer normalization, FFN, depth
Ready
->
7
Learning to Predict
Cross-entropy loss, backpropagation, batches, learning rate
Ready
->
8
Instruction Tuning & RLHF
From base model to assistant, reward models, PPO
Ready
->
9
GPT-2 from Scratch
Build and train a working GPT-2 in PyTorch
Ready
->
2
Pre-training
8 modules -- Training at scale: data, compute, and optimization
1
Overview
Word2Vec to GPT-3, the evolution of pre-training
Ready
->
2
Training Objectives & Architecture
CLM vs MLM, RoPE, RMSNorm, SwiGLU, FlashAttention
Ready
->
3
Scaling Laws & Optimization
Chinchilla, AdamW, learning rate schedules, gradient clipping
Ready
->
4
Training Data Engineering
CommonCrawl, deduplication, data mixing, quality filtering
Ready
->
5
Training Infrastructure
GPU clusters, distributed training, parallelism strategies
Ready
->
6
Advanced Objectives
Curriculum learning, synthetic data, multi-stage training
Ready
->
7
Evaluation During Pre-training
Benchmarks, validation loss, emergent capabilities
Ready
->
8
Case Study: LLaMA 3
How Meta built LLaMA 3 end-to-end
Ready
->
3
Post-training
5 modules -- From base model to helpful assistant
1
Overview
Why base models aren't enough, the alignment problem
Ready
->
2
Supervised Fine-Tuning
Instruction tuning, data quality, formatting
Ready
->
3
Preference Optimization
RLHF, DPO, reward models, PPO
Ready
->
4
Tools & Safety Tuning
Tool use, function calling, safety guardrails
Ready
->
5
Case Study: Tulu 3
Open post-training recipe end-to-end
Ready
->
4
Advanced Topics
3 modules -- Efficiency and specialization
1
Distillation
Compressing large models into smaller, faster ones
Coming Soon
2
LoRA
Low-rank adaptation for efficient fine-tuning
Coming Soon
3
Mixture of Experts
Sparse models, routing, conditional computation
Coming Soon