Semantic Search

Essential preprocessing technique that scales embedding vectors to unit length using L2 norm, ensuring consistent magnitude and making cosine similarity equivalent to dot product for faster computation.

Training technique enabling flexible embedding dimensions by learning representations where truncated vectors maintain good performance, achieving 75% cost savings when using smaller dimensions.

Maximum number of tokens an embedding model or LLM can process in a single input. Critical parameter for vector databases affecting chunk sizes, with modern models supporting 512 to 32,000+ tokens for long-document understanding.

Process of adapting pre-trained embedding models to specific domains or tasks for improved performance. Techniques include supervised fine-tuning, contrastive learning, and domain adaptation to optimize embeddings for particular use cases.

Number of components in an embedding vector, typically ranging from 128 to 4096 dimensions. Higher dimensions can capture more information but increase storage, computation, and costs. Critical design parameter for vector databases.