CLS Token in Vision Transformers
Learn how the CLS token acts as a global information aggregator in Vision Transformers, enabling whole-image classification through attention mechanisms.
8 min readConcept
Tagged with
attention
Explore machine learning concepts related to attention. Clear explanations and practical insights.
21
Concepts Found
Concepts Related to attention
Hierarchical Attention in Vision Transformers
Explore how hierarchical attention enables Vision Transformers (ViT) to process sequential data by encoding relative positions.
6 min readConcept
Multi-Head Attention in Vision Transformers
Explore how multi-head attention enables Vision Transformers (ViT) to process sequential data by encoding relative positions.
6 min readConcept
Positional Embeddings in Vision Transformers
Explore how positional embeddings enable Vision Transformers (ViT) to process sequential data by encoding relative positions.
5 min readConcept
Interactive Look: Self-Attention in Vision Transformers
Explore how self-attention enables Vision Transformers (ViT) to understand images by capturing global context, with CNN comparison.
6 min readConcept
ALiBi: Attention with Linear Biases
Learn ALiBi, the position encoding method that adds linear biases to attention scores for exceptional length extrapolation in transformers.
19 min readConcept
MHA vs GQA vs MQA: Choosing the Right Attention
Compare Multi-Head, Grouped-Query, and Multi-Query Attention mechanisms to understand their trade-offs and choose the optimal approach for your use case.
9 min readConcept
Attention Sinks: Stable Streaming LLMs
Learn about attention sinks, where LLMs concentrate attention on initial tokens, and how preserving them enables streaming inference.
17 min readConcept
Cross-Attention: Bridging Different Modalities
Understand cross-attention, the mechanism that enables transformers to align and fuse information from different sources, sequences, or modalities.
15 min readConcept
Grouped-Query Attention (GQA)
Learn how Grouped-Query Attention (GQA) balances Multi-Head quality with Multi-Query efficiency for faster LLM inference.
7 min readConcept
Linear Attention Approximations
Explore linear complexity attention mechanisms including Performer, Linformer, and other efficient transformers that scale to very long sequences.
6 min readConcept
Masked and Causal Attention
Learn how masked attention enables autoregressive generation and prevents information leakage in transformers and language models.
7 min readConcept
Multi-Query Attention (MQA)
Learn Multi-Query Attention (MQA), the optimization that shares keys and values across attention heads for massive memory savings.
7 min readConcept
Rotary Position Embeddings (RoPE)
Learn Rotary Position Embeddings (RoPE), the elegant position encoding using rotation matrices, powering LLaMA, Mistral, and modern LLMs.
8 min readConcept
Scaled Dot-Product Attention
Master scaled dot-product attention, the fundamental transformer building block. Learn why scaling is crucial for stable training.
6 min readConcept
Sliding Window Attention
Sliding Window Attention for long sequences: local context windows enable O(n) complexity, used in Mistral and Longformer models.
14 min readConcept
Sparse Attention Patterns
Explore sparse attention mechanisms that reduce quadratic complexity to linear or sub-quadratic, enabling efficient processing of long sequences.
7 min readConcept
Adaptive Tiling: Efficient Visual Token Generation
Learn adaptive tiling in vision transformers: dynamically partition images based on visual complexity to reduce token counts by up to 80% while preserving detail where it matters.
7 min readConcept
Prompt Influence Flow: How Instructions Propagate Through Model Layers
Deep dive into how different prompt components influence model behavior across transformer layers, from surface patterns to abstract reasoning.
6 min readConcept
Context Windows: The Memory Limits of LLMs
Interactive visualization of LLM context windows - sliding windows, expanding contexts, and attention patterns that define model memory limits.
6 min readConcept
Flash Attention: IO-Aware Exact Attention
Interactive Flash Attention visualization - the IO-aware algorithm achieving memory-efficient exact attention through tiling and kernel fusion.
7 min readConcept