Predicting Memorization Before Fine-Tuning

A classifier trained only on pre-fine-tuning base-model features predicts which sequences a fine-tuned model will memorize. The "Explore held-out examples" tab verifies this against ground truth on held-out examples from a fresh fine-tuning run; the "Score your own text" tab applies the same classifier to a passage you provide.

Pick a real Pythia-1.4B fine-tuning sequence and compare the classifier's pre-fine-tuning risk score with what the fine-tuned model actually generated.

Dataset
Filter

Datasets shown are public (FineWeb, PG-19, The Stack, OpenWebMath). The classifier uses only base-model features and was trained on a separate fine-tuning run.