The Art of the True Name
1
the intro
From Prompt to Response
Welcome
The Art of the True Name
1
the intro
The Runes Beneath Language
2
APIs and Model Serving
3
Tokenization during inference
4
Embeddings and Meaning Representation
The Labyrinth of Attention
5
Transformer Architecture
6
Attention Computation
7
KV Cache
8
GPU Execution
9
Quantization and Model Compression
10
Decoding and Sampling
The Forbidden Archive
11
Parametric Memory vs External Memory
12
Document Ingestion
13
Chunking Strategies
14
Embeddings for Search
15
Vector Databases
16
Retrieval Algorithms
17
Reranking
18
Context Construction
19
Grounded Generation
The Forging of Intelligence
20
Data Collection and Curation
21
Synthetic Data Generation
22
Tokenization During Training
23
Training Objectives
24
The Training Loop
25
Distributed Training
26
Distillation
27
Fine-Tuning and Parameter-Efficient Adaptation
28
Alignment and Preference Optimization
29
Reasoning Models and Test-Time Compute
30
Embedding Model Training
The Trial of the Answer
31
Human Evaluation
32
Benchmarks
33
Perplexity
34
Lexical Metrics
35
Semantic Metrics
36
Hallucination Metrics
37
RAG Evaluation
38
LLM-as-a-Judge
The Keepers of the Waking Engine
39
Tool Use and Function Calling
40
Agents and Multi-Agent Systems
The Art of the True Name
1
the intro
1
the intro
Welcome
2
APIs and Model Serving