The records behind this event
2-
AI Model
GPT-3
175 billion parameters and the discovery of in-context learning.
GPT-3 demonstrated that a large enough language model could perform tasks from a few examples in its prompt, no fine-tuning, launching the API business model and the scaling era. Trained on V100 clusters with 3.1×10²³ FLOP (disclosed), it is the ancestor of every product on this site's Models list.
Open record → -
Research Paper
Scaling Laws for Neural Language Models (2020)
Kaplan et al.: loss falls as a power law in compute: the investment thesis, quantified.
This OpenAI paper measured smooth power-law relationships between model performance and parameters, data and compute: the curves that turned 'bigger is better' from folklore into planning tool, and ultimately into capex justification for everything on this site.
Open record →