The theme of this week is content related to ChatGPT. At the beginning of March, OpenAI opened the ChatGPT API. Various related applications have emerged rapidly over the past two weeks.
Some of these products have become essential for my daily use. I am also thinking about how to utilize these tools to enhance my learning and work efficiency. Especially when using chatPDF to read articles or papers, continuously asking questions helps me stay more focused.
Products Used#
Articles Read#
Recently, I was translating a prompt engineering document. After reading this article, I finally understood why prompts are needed to control the output of ChatGPT.
Principle
A generative model is a function that can take a structured collection of symbols as input and produce a related structured collection of symbols as output
Deterministic VS Stochastic
Relationship matters
- The diversity and complexity of relationships between symbols
- Latent space -> A multidimensional space of possible output results
Probability distributions
- What methods can change the probability distribution of LLM models
- Training
- Fine-tuning
- Reinforcement Learning from Human Feedback (RLHF)
Some Thoughts#
Techniques Learned#
- paul-graham-gpt
- Using embedding for text search => Using cosSim to calculate similarity
export const cosSim = (A: number[], B: number[] ) =>
{
let dotproduct = 0
let MA = 0
let MB = 0
for (let i = 0; i < A.length; i++) {
dotproduct += A[i] * B[i]
MA += A[i] * A[i]
MB += B[i] * B[i]
}
MA = Math.sqrt(MA)
MB = Math.sqrt(MB)
const similarity = dotproduct / (MA * MB)
return similarity
}
- openai-translator
- Using Tauri to package web applications into desktop apps
Opportunities for Independent Developers#
- The developer of typingmind earned $20,000 in one week