3 ms·
Could you provide more details on this matter? Specifically, I'm interested in knowing which base model you've utilized and the approach you've taken to fine-tu
by alwinaugustin 3y ago
Could you provide more details on this matter? Specifically, I'm interested in knowing which base model you've utilized and the approach you've taken to fine-tune it. Your insights would be greatly appreciated and highly beneficial.
- m3kw9 3y agoFor narrow stuff you can do better job than base gpt4/mistral/etc model. You fine tune it with your very custom data, stuff that got didn’t seem to be trained on, it will generalize it well.
- simonw 3y agoHave you done this? How did you do it? I've been looking forward to someone providing a detailed guide on how to "fine tune it with your custom data" for ages!
- arbitrandomuser 3y agothis is imo the secret sauce that gives people an edge and not a lot of people will want to reveal
- washadjeffmad 3y agoYou're not wrong. There's been a lot of drama over licensing and releasing datasets, and a lot of the LLM scene are just pitchmen and promoters with no better grasp over what they're doing than "trust me, it's better". Like with "prompt engineering", a lot of people are just hiding how much of the heavy lifting is from base models and a fluke of the merge. The past few "secret" set leaks were low/no delta diffs to common releases. I said it a year ago, but if we want to wowed, make this a job for MLIS holders and references librarians. Without thorough, thoughtful curation, these things are just toys in the wrong hands.
- ipaddr 3y agohttps://www.datacamp.com/tutorial/fine-tuning-llama-2 https://www.datacamp.com/tutorial/fine-tuning-llama-2
- anon373839 3y agoThis is a very nice resource: https://github.com/mlabonne/llm-course https://github.com/mlabonne/llm-course