PinnedInThe StartupbyHarshit Dawar·Nov 25, 2020Probability VS LikelihoodThis blog aims to explain the difference between the Probability & the Likelihood. This topic is very important to understand, but the…A response icon9A response icon9
InTowards AIbyHarshit Dawar·5d agoLoRA Fine-Tuning: Match Full Fine-Tuning With Far Less GPU (with Hand Drawn Explanations)!For practitioners who have hit ‘CUDA out of memory’: why LoRA works, the settings that matter, when to pick DoRA, rsLoRA, or PiSSA instead…A response icon1A response icon1
InTowards AIbyHarshit Dawar·Nov 8Agentic/Feedback Loop vs Chatbot: What Actually Makes an AI an Agent!A plain-English guide for everyone: the four-step cycle of AI Agents, self correceting loop, a working Python example, and where it breaks…
InTowards AIbyHarshit Dawar·Nov 5Catastrophic Forgetting in LLMs & Its Solutions: Fine-Tune Without Losing Skills!Replay, LoRA, EWC, KL anchoring, merging, and RAG explained with Python code, so you can fine-tune on your data without regressions.A response icon1A response icon1
InTowards AIbyHarshit Dawar·Nov 3MAP@k vs Precision@k vs MRR: Building Capability to choose the right one for each situationClarifying the confusion once and for all in evaluating RAG and many other use cases!A response icon1A response icon1
InArtificial Intelligence in Plain EnglishbyHarshit Dawar·Nov 1Sequence 2 sequence Models vs GPT Models — Demystifying the core and the myths!Understanding the most important difference between 2 similar looking things!A response icon1A response icon1
InTowards AIbyHarshit Dawar·Oct 31MRR vs Recall@k vs Precision@kWhen to use which, with their pros and cons!A response icon1A response icon1
InArtificial Intelligence in Plain EnglishbyHarshit Dawar·Oct 29What is Window Attention in LLMs?The trick that enables LLMs to read long text by looking through a small sliding window!A response icon1A response icon1
InTowards AIbyHarshit Dawar·Oct 25What is Attention Sink in LLMs — The necessary evil that keeps the balance?The behaviour of focusing on the first few tokens when there is nothing meaningful with LLMs!A response icon1A response icon1
InTowards AIbyHarshit Dawar·Oct 22Multi Query Attention vs Grouped Query Attention — Understanding the perfect use case for each and…This is behind the performance of your LLMs!A response icon1A response icon1