Large Language Models
Scaling, alignment, prompting, retrieval and serving — how LLMs actually work
25 cards · first 3 free · flip each card to test yourself
04
The KV Cache
PixelBank Premium
05
The Modern Decoder Stack
PixelBank Premium
06
Grouped-Query Attention
PixelBank Premium
07
Mixture of Experts
PixelBank Premium
08
Pretraining Data
PixelBank Premium
09
Distributed Training
PixelBank Premium
10
In-Context Learning
PixelBank Premium
11
Instruction Tuning
PixelBank Premium
12
LoRA
PixelBank Premium
13
QLoRA & Quantized Fine-Tuning
PixelBank Premium
14
Reward Modelling
PixelBank Premium
15
RLHF with PPO
PixelBank Premium
16
Direct Preference Optimization
PixelBank Premium
17
Prompt Engineering
PixelBank Premium
18
Chain-of-Thought
PixelBank Premium
19
Structured Output
PixelBank Premium
20
Embeddings & Vector Search
PixelBank Premium
21
Chunking
PixelBank Premium
22
RAG Pipelines
PixelBank Premium
23
Tool Use & Function Calling
PixelBank Premium
24
Hallucination & Guardrails
PixelBank Premium
25
Quantization & Serving
PixelBank Premium