QED-Nano: Teaching a Tiny Model to Prove Hard Theorems
📝
74
Who needs 1T parameters? Olympiad proofs with a 4B model
ruins
ruinnight
AI & ML interests
None yet
Organizations
None yet
liked a
Space 4 months ago
liked 3
Spaces 7 months ago
Unlocking On-Policy Distillation for Any Model Family
📝
109
Visualize on‑policy distillation token alignment
Scaling FineWeb to 1000+ languages: Step 1: finding signal in 100s of evaluation tasks
📝
94
Evaluate multilingual models using FineTasks
The Smol Training Playbook
📚
3.2k
The secrets to building world-class LLMs
liked 3
models 9 months ago
liked a
dataset 9 months ago
liked a
dataset about 1 year ago