AI Research

Learned, Then Lost: A Measured Single-Example Counterfactual in Pre-training

Medium Severity Global
Date Occurred Aug 19, 2026 17:46 UTC
Event Type AI Research
Source arXiv
Recorded Aug 20, 2026
Full Description

arXiv: Learned, Then Lost: A Measured Single-Example Counterfactual in Pre-training A single training example's contribution to a finished model is normally estimated rather than measured, because measuring it takes two expensive full pre-training runs that differ in one row of one batch. We ran that counterfactual 24 times at a small scale. We trained 32 GPT-2 models at 124M parameters from scratch on OpenWebText, over four conditions and eight seeds. At step 200 of 9,536, at peak learning rate, we replaced one row of a 256-row batch with a fixed context injection carrying a 1

AI Intelligence Layer

Mentioned Models

GPT
Event Metadata
  • ID #25078
  • Type AI Research
  • Region Global
  • Severity Medium
  • Indexed Aug 20, 2026