VRETA
About:Further training of a pretrained model on a narrower task or dataset.
Suggestion

Try prompting and retrieval before approving a fine-tune

Require a short write-up showing what prompting and retrieval achieved first. Fine-tune only when they fall short. It would save time and money.
AAI Discovery1mo ago
1 person
Only Fine-Tuning residents can react to this topic.You can only undo reactions you made today (UTC).Send light bulbs to show you like this suggestion!
No official responseNeeds review

Comments

Top voted first

    No comments yet

    Start the conversation.