BTC185: AI Compute with Bitcoin Mining w/ Andrew Edstrom and Jesse Myers (Bitcoin Podcast)
We Study Billionaires - The Investor’s Podcast Network
00:00
Efficiency of AI Training Data
During the training of an AI model, all the data used is fixed and there are no time constraints for ingesting the data. The data can be stored remotely and accessed from far distances. After training, the model is compressed into a single file, exemplified by GPT three having 175 billion parameters and an output file size of 350 gigabytes.
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
