Rishi Kumar·Sep 29RAG Retrieval: Vector Search, Hybrid Search, Bi-Encoders, ANN, HNSW, IVF and Evaluation MetricsIn the previous article, we explored the fundamentals of RAG.
Rishi Kumar·Sep 19RAG Fundamentals: Embeddings, Chunk Size, Chunk Overlap, and Retrieval QualityHow can an LLM answer questions using information that was not part of its training data?
Rishi Kumar·Aug 8How Are Large Language Models Trained at Scale?In the previous articles, we explored how Large Language Models are trained, how scaling laws work, and why model size, training data, and…
Rishi Kumar·Aug 2Why Do Larger Language Models Perform Better?In the previous article, we explored how a Large Language Model learns using:
Rishi Kumar·Jul 27How Does a Large Language Model Actually Learn?In the previous articles of this series, we explored how Transformers work, how Large Language Models generate text, and the optimization…
Rishi Kumar·Jul 4Inference Optimization in Large Language ModelsIn the previous articles, we learned how Large Language Models are built and how they generate text.
Rishi Kumar·Jun 27How Large Language Models Generate Text: Greedy Search, Beam Search, Top-K, Top-P, and TemperatureBy now, we’ve explored how Large Language Models are built, from Transformers to Mixture of Experts (MoE).
Rishi Kumar·Jun 19Understanding Mixture of Experts (MoE): How Modern LLMs Scale to Trillions of ParametersIn the previous articles, we explored:A response icon1A response icon1
Rishi Kumar·Jun 6Understanding the Three Categories of Transformer ArchitecturesIn the previous articles, we explored several important concepts behind modern Transformers:
Rishi Kumar·May 30Improving Transformer Efficiency: Attention Approximation, MHA, MQA, and GQAIn the previous article, we explored how Transformer architectures improved positional understanding using: