Inner Workings

How machine learning and large language models work underneath: training, embeddings, transformers and fine-tuning.

Notes in this section

• Embeddings — Embeddings are a fundamental concept in AI models, particularly in natural language processing (NLP) and computer vision.
• Fine tuning — Fine-tuning is the process of taking a pre-trained model and training it further on a smaller, specific dataset to adapt it to a…
• The Road to ChatGPT — Architectural Evolution of Large Language Models — The complete technical evolution from N-grams to ChatGPT — Transformers, GPT-1 through GPT-4, InstructGPT, and the three-stage…
• Overfitting vs underfitting — Two fundamental concepts in machine learning.
• The Data Science Lifecycle & CRISP-DM Process — The 9 phases of the data science and machine learning lifecycle — problem formulation, EDA, feature engineering, model…
• Reinforcement Learning with Human Feedback — Reinforcement Learning from Human Feedback (RLHF) is a technique used to fine-tune machine learning models, particularly large…
• Supervised vs Unsupervised Learning — In supervised learning, you train the model with a set of input data and a corresponding set of paired labeled output data. The…
• Whats a T in GPT? — A transformer is a type of neural network architecture that’s particularly well-suited for natural language processing (NLP)…