LLM training dynamics & data influence
What is LLM training dynamics & data influence?
This trend covers the techniques and processes that shape how language models learn from data and get modified after training, including fine-tuning methods, reinforcement learning approaches, and ways to influence model behavior without retraining from scratch.
Understanding how models are trained and adapted is critical for evaluating their reliability, controlling their behavior, and assessing whether improvements require expensive retraining or can be achieved through cheaper modifications.
References
- Understand LLaMA: How Meta Democratised Large Language Models — Medium: Large Language Models
- Understanding Large Language Models, Their Functionality and Future — Medium: LLM
- Introduction to LLMs (Large Language Models) — Medium: Large Language Models
- LoRA: Small Updates, Big Possibilities - A Practical Guide to Efficient LLM Fine-Tuning — Medium: LLM
- Baking a Model: A Metaphor for LLM Training — Hacker News
- Model Genome: Fingerprinting Whether an LLM Was Trained from Scratch or Derived — Hacker News
- Reflexion: Train an LLM Agent Without Touching Its Weights — YouTube
- The Modern LLM Stack: Why Model Weights Aren't Enough. — YouTube