See this notebook
- Note on LLMS
- LLM use cases
- Proprietary services
- Open source models
- Sentence transformers
- Smeantic Search
- Approx nearest neighbour search
- Retireve and Re-rank
- Retieval: Bi-Encoder 10.Re-Ranker: Cross-Encoder
- How FAISS works
- What is vector store?
- How are vector databases used?
- What are the benefits of vector databases?
- What is LLM Prompter?
- What is quantization?
- Why GGUF format?
- Complete chatbot cycle? Notebooks and other material on LLMs
- NLP Moore's Law
- Resources
- Prompt Engineering
Six strategies for getting better results
Chain of thought
Example of complete prompt
What is Zero-Shot Chain-of-Thought prompting?
4.Prompts for general coding workflows
5.What is RAGWhat is the difference between Retrieval-Augmented Generation and semantic search?
1.A list/collection of open source and/or local AI tools and solutions.
2.Byte Pair Encoding
3.What is Ollama?Models supported by ollama. See List here
Create custom ollama models by using Modelfile and also see this Modelfile on github
Ollama based webUIs
Where are ollama models stored?
4.What is ollama-index
Why LlamaIndex for Context Augmentation?
5.Nomic Text Embeddings 6.How are LLM models named
7.Types of LLM modelsUncensored models--What are they? with examples
Why self-supervised learning
Distortion
Rotation
Patches
2.langchain
Why is langchain important? langchain use cases with example code
How does langchain work?
What are core components of langchain?
LLM Parameters (temperature, top-k etc)
langchain QuickStart
langchain API Reference langchain video lectures
langchain simple tutorials
Broad steps for RAG
How is langchain organized
Installation
Wrining Prompts
How to create a prompt
Chains in langchain
LLM Chain
Sequential Chain
Retrieval chain
Getting data from the Web
FAISS
Generate Answers to Questions using LLM
Conversational Retrieval Chain
Reading pdf files: from disk from folder/directory
Simple RAG for GitHub issues using Hugging Face Zephyr and LangChain Link
What is a RAG
RAG on Colab using Huggingface models and langchain
Creating and using Quantized models
Using FAISS library
Design prompt template
HuggingFace pipeline and langchain pipeline wrapper