02-05-2024 04:45 PM
Generative AI Fundamentals Part 5 - Fine-tuning:
60 second version: https://youtu.be/oEbOJiYIRxE?si=7u0PTFrWKEAiHgvH
What is Fine-tuning?
Fine-tuning involves tailoring a pre-trained model to perform specific tasks.
While you could do unsupervised fine-tuning, in most cases it is a supervised learning process where you use a dataset of labeled examples with prompts and responses to update the weights of an LLM & make the model improve its ability for specific tasks such as translation, summarization, writing articles etc.
Once a model has been fine-tuned, you won't need to provide as many examples in the prompt. This saves costs and enables lower-latency requests.
How does Fine-tuning work?
At a high level, fine-tuning involves the following steps:
- Prepare and upload training data
- Train a new fine-tuned model
- Evaluate results and go back to step 1 if needed
- Use your fine-tuned model
Detailed explanation: https://youtu.be/7Qkzn6r5H1M?si=UNSbK-fFeCSlO6zd
In the next video, we will dive into the hot topic of RAG or Retrieval-Augmented Generation. Thanks for tuning in