LLM Pre-training vs. Instruction-Tuning -LLaMa 65B pre-trained -Only simple fine-tuning, w/ only 1k (carefully chosen) data points, no RLHF -Can plan trips & speculate about alternate histories -Generalizes to unseen tasks -Humans prefer it over GPT-3 https://t.co/6edcWjVz04 https://t.co/wQvbbxpmtc
— John Nay (@johnjnay) May 22, 2023
from Twitter https://twitter.com/johnjnay
May 21, 2023 at 07:06PM
via IFTTT
No comments:
Post a Comment