4 ms·
> create a fine tuned version of GPT No. It's unclear fine-tuning is happening. Many are guessing it's RAG.
by nomel 3y ago
> create a fine tuned version of GPT
No. It's unclear fine-tuning is happening. Many are guessing it's RAG.
- vunderba 3y agoMy bad, I wasn't intending fine tune to mean fine tuning the model itself. And leveraging vectorized embeddings as I mentioned IS FUNCTIONALLY a rag-style architecture.