Pith. sign in

REVIEW 4 cited by

Emergent and Predictable Memorization in Large Language Models

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2304.11158 v2 pith:VLY6BVOL submitted 2023-04-21 cs.CL

Stella Biderman , USVSN Sai Prashanth , Lintang Sutawika , Hailey Schoelkopf , Quentin Anthony , Shivanshu Purohit , Edward Raff This is my paper · ORCID
classification cs.CL
keywords memorizationmodelmodelsdatalanguagelargepythiasequences
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Memorization, or the tendency of large language models (LLMs) to output entire sequences from their training data verbatim, is a key concern for safely deploying language models. In particular, it is vital to minimize a model's memorization of sensitive datapoints such as those containing personal identifiable information (PII). The prevalence of such undesirable memorization can pose issues for model trainers, and may even require discarding an otherwise functional model. We therefore seek to predict which sequences will be memorized before a large model's full train-time by extrapolating the memorization behavior of lower-compute trial runs. We measure memorization of the Pythia model suite and plot scaling laws for forecasting memorization, allowing us to provide equi-compute recommendations to maximize the reliability (recall) of such predictions. We additionally provide further novel discoveries on the distribution of memorization scores across models and data. We release all code and data necessary to reproduce the results in this paper at https://github.com/EleutherAI/pythia

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 4 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Bits and Memories: Measuring Verbatim Extraction Across LLM Quantization

    cs.LG 2026-07 conditional novelty 7.0 of 10

    Quantizing LLMs selectively forgets memorized text faster than capability, but 1B-scale 4-bit models still extract ~72% of memorized sequences, so quantization is not a privacy defense.

  2. Leak It: A Probabilistic Approach to Training-Data Extraction from Black-Box Language Models

    cs.LG 2026-07 conditional novelty 6.0 of 10

    Sampling-based LLM attacks reproduce exact identifiers from 16.6% of 500 Pile documents at Pythia-6.9B even though aggregate sampling-MIA adds no signal over blind baselines, so privacy audits should report per-docume...

  3. Infrared Organization and Critical Cognitive Field Formation in Transformer Dynamics

    cs.LG 2026-07 reject novelty 6.0 of 10

    Slow relaxation modes in Pythia transformers accumulate toward zero rate during training, yielding a near-flat infrared spectrum and 1/t memory kernels—but the claimed 'critical cognitive field formation' is not direc...

  4. Code Simulation as a Proxy for High-order Tasks in Large Language Models

    cs.LG 2025-02 conditional novelty 5.0 of 10

    LLM performance on naturalistic reasoning tasks tracks performance on equivalent Python code simulation, but the effect is partly driven by pattern matching and memorization rather than faithful execution.

Pith tools