NNishanthan KarunakaranOct 15 · 6 min readUnderstanding Transformer: The Architecture Behind Modern AIWhat Is a Transformer? A Transformer is a neural network architecture introduced by Google researchers in 2017 in the paper Attention Is All You Need. It uses attention mechanisms to help the model le1nishanthan.hashnode.dev
EEkansh NangiaSep 17 · 18 min readIntroduction To Mechanistic InterpretabilityEpistemology Epistemology is the branch of philosophy thats concerned with the nature and limits of knowledge. What knowledge actually is, how we get it, and how we tell the difference between somethiekanshn.hashnode.dev
SSunney SoodSep 11 · 26 min readHow Large Language Models Actually Work: From Your Prompt to Every Word They WriteA complete, plain-English mental model — from tokens to attention to the exact moment ChatGPT picks its next word. When you type a question into ChatGPT and an answer appears a few seconds later, itsunneysood.hashnode.dev
RRahul Sai Indeevar VSep 7 · 15 min readAdvanced Transformers: Positional Encodings (RoPE) & Efficient Attention Mechanisms (KV Caching, MQA, GQA, MLA)In modern Large Language Model (LLM) architectures, two foundational engineering bottlenecks dictate model scalability and inference latency: Representing Order: How self-attention mechanisms—which a13Arahul-ai.hashnode.dev
RRahul Sai Indeevar VSep 2 · 11 min readClassical NLP to Attention: Guide to Sequence ModelingNatural Language Processing (NLP) has undergone a massive paradigm shift over the past decade. We have moved from handcrafted rule-based systems and statistical frequencies to complex sequential archi1rahul-ai.hashnode.dev
VVijay Ram EnagantiAug 31 · 9 min readReproducing Sequential AttentionTL;DR Implemented Google's Sequential Attention feature-selection algorithm (ICLR 2023) from scratch in PyTorch (seq-attention/), faithful to Algorithm 1: a softmax mask over per-feature attention lo1vijay-ram.hashnode.dev
VViv ChaudharyAug 22 · 12 min readFive Mental Models That Make Modern AI Click Most people try to learn AI by collecting definitions. They pick up "transformer," "embedding," "context window," and "fine-tuning" the way you'd collect vocabulary flashcards, one term at a time, eactruenorthbits.hashnode.dev
AAkashAug 5 · 7 min readAbstractive Text Summarization with Attention Mechanism: Seq2Seq Model for Academic PapersSummarizing long academic papers or documents manually is tedious and time-consuming. We built an Abstractive Text Summarization System that automatically generates concise, meaningful summaries from akashbuilds.hashnode.dev
PPiyush JadhavAug 3 · 4 min readAttention Mechanism in Deep Learning: The Foundation of TransformersIntroduction Artificial Intelligence (AI) has transformed the way machines understand language, recognize images, and make intelligent decisions. One of the most significant breakthroughs in deep learattentionmechanism.hashnode.dev
AAman JainJul 21 · 7 min readHow Much Do Attention Heads and Activation Functions Really Matter?TL;DR I tested two architectural choices in a small GPT model while keeping the rest of the setup stable: Attention heads: 1, 4, and 8 heads under fixed-token and fixed-time budgets. Activation funccurious-pm.hashnode.dev
