Skip to content

Paper 040 / Anthropic

Discovering Language Model Behaviors with Model-Written Evaluations

Model-written evaluations use language models to propose test cases for behaviors such as bias, deception, power seeking, and other difficult-to-enumerate tendencies.

UPDATED 16 Jul 2026SOURCE-LED REVIEWRESEARCH REVIEW
01

Paper, researchers, and primary source

Major lab research / evaluation

Model-written evaluations use language models to propose test cases for behaviors such as bias, deception, power seeking, and other difficult-to-enumerate tendencies.

CONTRIBUTION / 01

Contribution 1

Created a pipeline for generating behavioral evaluations with language models.

CONTRIBUTION / 02

Contribution 2

Released datasets spanning many hypothesized model behaviors.

CONTRIBUTION / 03

Contribution 3

Compared behavior across model scale and training interventions.

02

Research context

evaluation / 2022

Discovering Language Model Behaviors with Model-Written Evaluations places model written evals inside the broader evaluation discussion at Anthropic, with behavioral evaluation supplying a second analytical lens. Read together, the source records three advances: Created a pipeline for generating behavioral evaluations with language models; Released datasets spanning many hypothesized model behaviors; and Compared behavior across model scale and training interventions. Keeping those moves together prevents safety from being detached from its evidence. For an implementation review, the relevant consequence is that synthetic evaluations can broaden coverage quickly, but construct validity and human auditing determine whether they measure the intended behavior.

03

Methods and evidence reading

1 cataloged method notes

METHOD / 01

Method 1

The experimental design in Discovering Language Model Behaviors with Model-Written Evaluations tests created a pipeline for generating behavioral evaluations with language models and released datasets spanning many hypothesized model behaviors against the paper's documented baselines, datasets, model variants, or systems workloads.

How to read the evidence

The evidentiary value of Discovering Language Model Behaviors with Model-Written Evaluations comes from the relationship among its reported moves. Two entries define the method-level claim: Created a pipeline for generating behavioral evaluations with language models; and Released datasets spanning many hypothesized model behaviors. The cataloged result is: Compared behavior across model scale and training interventions. On that basis, synthetic evaluations can broaden coverage quickly, but construct validity and human auditing determine whether they measure the intended behavior. The catalog nevertheless records that the demonstrated model written evals result belongs to a setup defined by rater instructions, model revisions, evaluator models, construct validity, selected threat model, and prompt sampling, not to every later behavioral evaluation system. Reproduction work should separate genuine model written evals transfer from behavior caused by a changed behavioral evaluation setup. A reproduction ledger for Discovering Language Model Behaviors with Model-Written Evaluations should preserve model written evals, vary behavioral evaluation, and retain a counterexample tied to Released datasets spanning many hypothesized model behaviors before judging transfer.

04

Findings in the source record

1 paper-specific findings

  1. The reported evidence in Discovering Language Model Behaviors with Model-Written Evaluations supports compared behavior across model scale and training interventions.
05

Practical implication for AI builders

Anthropic / 2022

06

Proposed BrokenGPT application

Research blueprint / proposed status

INTEGRATION POINT / 01

Proposed route placement / model written evals

For a proposed BrokenGPT experiment based on Discovering Language Model Behaviors with Model-Written Evaluations, let evaluator models draft new behavioral tests, then require human validation, deduplication, and documented pass criteria before use in release gates. Keep the model written evals path isolated, versioned, and attributable to this research record.

VALIDATION METRIC / 02

Proposed acceptance test / behavioral evaluation

Validate the proposed model written evals route against the paper's reported outcome: Compared behavior across model scale and training interventions. Measure evaluator agreement, slice stability, metric coverage, and calibration for the Discovering Language Model Behaviors with Model-Written Evaluations candidate, then isolate behavioral evaluation regressions before judging the proposed model written evals route.

TRADEOFF / 03

Proposed decision boundary / safety

Balance breadth, repeatability, and construct validity before promoting the proposed safety design. Because reusing the mechanism calls for separate evidence about deployment drift, adversarial adaptation, language coverage, unsampled behaviors, and judge bias, not an inference from the original benchmark alone, adoption remains conditional on replay under BrokenGPT's selected model, runtime, and policy configuration.

07

Limitations, verification, and source

Boundaries recorded with the paper

Limitations

  • The demonstrated model written evals result belongs to a setup defined by rater instructions, model revisions, evaluator models, construct validity, selected threat model, and prompt sampling, not to every later behavioral evaluation system.
  • Reusing the mechanism calls for separate evidence about deployment drift, adversarial adaptation, language coverage, unsampled behaviors, and judge bias, not an inference from the original benchmark alone.

PRIMARY SOURCES

  1. 01
    Discovering Language Model Behaviors with Model-Written Evaluations

    Anthropic — Primary primary arXiv paper / 19 December 2022 / Ethan Perez, Sam Ringer, Kamilė Lukošiūtė, and 60 more

Related research reviews

View all 100 credited research papers

STRAIGHT ANSWERS

Frequently asked questions

01What does Discovering Language Model Behaviors with Model-Written Evaluations study?

Model-written evaluations use language models to propose test cases for behaviors such as bias, deception, power seeking, and other difficult-to-enumerate tendencies.

02Which methods does Discovering Language Model Behaviors with Model-Written Evaluations use?

The experimental design in Discovering Language Model Behaviors with Model-Written Evaluations tests created a pipeline for generating behavioral evaluations with language models and released datasets spanning many hypothesized model behaviors against the paper's documented baselines, datasets, model variants, or systems workloads.

03What does Discovering Language Model Behaviors with Model-Written Evaluations report?

The reported evidence in Discovering Language Model Behaviors with Model-Written Evaluations supports compared behavior across model scale and training interventions.

04What is the proposed BrokenGPT application for Discovering Language Model Behaviors with Model-Written Evaluations?

Proposed: let evaluator models draft new behavioral tests, then require human validation, deduplication, and documented pass criteria before use in release gates.

MAJOR LAB RESEARCH / PAPER 040

Continue after Discovering Language Model Behaviors with Model-Written Evaluations

After Discovering Language Model Behaviors with Model-Written Evaluations, browse the full index for adjacent evaluation research and work from Anthropic.

Open the paper index