2 ms·
100B, 220B, and 600B LLM models on HuggingFace
- squigz 3y agoI'm not sure I like this trend of linking to reddit/Twitter/X/whatever when one could link directly to the source... https://huggingface.co/deepnight-research https://huggingface.co/deepnight-research
- behnamoh 3y agothink of it as a data point that’s richer because it’s wrapped in a context and includes meta data as well (I’ve been doing Pydantic recently…)