Scaling Compute on Context — Jack Morris, Engram
Morris trains a model on 10,000 financial reports to 0.00001 loss — perfect document recall — then shows generation collapsing, and reframes the fix along the three axes behind deep learning.
Featured · Jack Morris
1 source
Daily brief
Get tomorrow's AI brief in your inbox
More stories today
- OpenAI-backed Thrive Holdings raises $2B to bring AI to the enterprise
- Open Instruct tutorial covers LLM post-training with SFT, DPO, GRPO
- Twitch streamers can now opt out from training Amazon's AI
- OpenWALDO project launches to create shared, open-source AI training dataset
- MIT Technology Review report: Legacy data systems limit AI agents