This repository consist of the projects that are implemented by Python and LangChain framework during my research on LLMs.
In this research, firstly I focused on how LLMs work and their architecture. After understanding these topics, I focused on the prompting tecniques and the RAG technique.
- How LLMs Work
- Transformer Architecture
- Pre-Tranining and Fine-Tuning
- Tokenization
- Vector Embeddings
- LLM Evaluation Metrics
- Zero-Shot, One-Shot and Few-Shot Techniques
- Chain of Thought Technique
- Self Consistency Prompting
- Persona Based Promting
- RAG Technique
After understanding how the LLMs work, I developed 2 simple projects using LangChain framework and Python.