Generative AI

Avatar
Praveen Yarlagadda 10 Months Ago
9 Likes
    • 1.1. Precision & recall with an example
    • 1.2. Evaluating a model using Accuracy
    • 1.3. Precision & recall trade off in binary classification with an example
    • 2.1. Experiments, outcomes, events, probabilities #math #probability
    • 2.2. Mutually exclusive events #probability #math
    • 2.3. Independent events #math #probability
    • 2.4. Conditional probability - main concept behind generative AI. #math #probability
    • 2.5. Bayes Theorem - basis for Generative AI #genai #math #probability #stats #bayes
    • 3.1. Gradient of a function with single variable - basis for training neural nets #ne
    • 3.2. Gradient of a function with multiple variables #math #genai #neuralnet #chatgpt
    • 3.3. Mathematical model of a neuron. Basis for all LLMs like #chatgpt #claude #anthro
    • 3.4. Neuron vs neural network. An axon terminal contains various neurotransmitters th
    • 3.5. A neuron with a data point that has two features - X1 & X2. Expected output is Y
    • 3.6. Neural network forward pass in action with two data points each with two feature
    • 3.7. Back propagation example - #neuron #ai #genai #training #calculus
    • 4.1. Transformer architecture from 'Attention is all you need' paper.
    • 4.2. How encoders function in transformer models. #chatgpt #mistral #transformer #llm
    • 4.3. How decoders function in transformer models #chatgpt #mistral #transformer #llms
    • 5.1. Hyper parameters of Nano GPT (simplified GPT). #nanogpt #chatgpt #genai
    • 5.2. Input to Nano GPT. How generated labeled data is organized and fed to Nano GPT.
    • 5.3. What is self-attention in transformers. #nanogpt #chatgpt #genai
    • 5.4. How information is aggregated at each token in a sentence 'Sunshine Brings Hope'
    • 5.5. How dimensions of input tensors change as they pass through different layers of
Avatar

Precision & recall with an example

Chapters
1. Metrics
2. Probability
3. Neural Network
4. Transformers
5. NanoGPT