2Dataset streaming for distributed SOTA model training now more efficient (opens in new tab)(huggingface.co)2lhoestq5mo ago0
4A prompt is worth 1000 data points: combining GPT3 prompting and fine-tuning (opens in new tab)(huggingface.co)3lhoestq4y ago0
5Datasets: Release 1.3 brings dataset versioning, on-the-fly transforms and more (opens in new tab)(reddit.com)2lhoestq5y ago0
