LLMS Prompts (Q&As and More)

ChatGPT (4o mini)

Code

Vanilla Python with/without NumPy

• Q: generate a perceptron in plain vanilla python code with numpy
  • note - generates a single-layer perceptron training on logicial xor, see Q: explain if you can train with success a single-layer perceptron on logical xor
• Q: generate a perceptron in plain vanilla python code with numpy and train on logical or
  • Q: try the code without numpy

• Q: generate a multi-layer perceptron in plain vanilla python code with numpy and train on logical xor
  • Q: try the code without numpy

With PyTorch

• Q: generate a perceptron in python code with pytorch
  • note - generates a single-layer perceptron training on logicial xor, see Q: explain if you can train with success a single-layer perceptron on logical xor
• Q: generate a perceptron in python code with pytorch and train on logical or

• Q: generate a multi-layer perceptron in python code with pytorch and train on logical xor
  • Q: change the neural network to use 3 hidden neurons and 2 outputs

Theory

• Q: what is a single-layer perceptron?
• Q: explain if you can train with success a single-layer perceptron on logical xor
• Q: what is the best multi-layer perceptron configuration to train with success a perceptron on logical xor
  • note - includes running sample code in python with pytorch!
• Q: what's the difference between gpt and bert neural networks?
• Q: can you explain the word2vec neural network model?
• Q: what are some options for positional encoding of word embeddings?
• Q: what are the pro and cons of byte pair encoding tokenizers compared to alternatives?
• Q: can you explain transformers with attention mechanisms?
• Q: what differences are there in the neural network layers of gpt 2 and llama 3?
• Q: what are the memory requirements for gpt-2?
• Q: how is the gpt-3 neural network architecture different from gpt-2?
• Q: can large language model think?

History

• Q: timeline of a.i. research
• Q: timeline of a.i. research in neural networks
• Q: timeline of a.i. research in neural networks with transformers and attention
• Q: timeline of decoder-only transformer neural network models

that's all for today. thank you.

You're very welcome! I'm glad I could help. If you have more questions in the future, don't hesitate to reach out. Have a great day!