Concept
Fine-tuning
Further training an existing model on a smaller, specific dataset to specialize its behavior.
Fine-tuning continues training a pre-trained model on a narrower dataset to specialize it: a domain, a task, a style or a safety behavior. Open-weight models make fine-tuning available to anyone, which cuts both ways: Microsoft Decision-1 is a fine-tune of Qwen, while Cantina’s “abliterated” Apex Flash-1 variant shows fine-tuning can also strip refusals.
Connections
Fine-tuning and everything one link away. Drag to move, scroll to zoom, click a node to open it.Open in the full graph →
What links here
Company Cantina
Concept Large language model (LLM) · Open-weight vs. closed models
Sources
- Wikipedia, “Fine-tuning (deep learning)”