HN Explorer

Ask HN: What do you find most difficult about fine tuning AI models?

1 points by Nevin1901 ↗ HN
Those of you who have fine tuned NLP (GPT-X/Llama) or diffusion models, what was the most difficult or annoying part about the process?

I'm trying to help solve problems in this field, so I'm curious about your experiences.

1 comment

[ 2.7 ms ] story [ 12.3 ms ] thread
tlack
On the LLM side: the cost / amount of resources that need to be brought to bear for long periods of time

On the Stable Diffusion side: labeling, best practices, file management, quality control