I want to try out AI/ML finetuning on a laptop.
Nothing too involved. I just want to see the output of the model improve in a couple of hours as I finetune it.
Is there a pre-trained model out there that is small enough to do this? (Ideally something that can be installed via a single Docker-Compose command.)
- Just fine tune one of OpenAIs models. Try doing it with GPT 4 and use outputs from GPT o1/o3 to produce data.
- Hi, depends on what kind of model you want to fine-tune. Could you be a little more specific?