# Ripple effect magnitude prediction (ℓ2 logit shift) on Curated ripple-effect corpus 1.0 (test) Source: [Ripple effect magnitude prediction (ℓ2 logit shift) on Curated ripple-effect corpus 1.0 (test)]() Collection index: [llms.txt]() Data updated: 2026-03-23T11:04:17.714Z Evaluation of model performance on predicting ripple effect magnitude using the test split of the Curated ripple-effect corpus 1.0, measured by Spearman correlation (ρs). A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Ripple effect magnitude prediction (ℓ2 logit shift) | | Dataset (including any stated split) | Curated ripple-effect corpus 1.0 (test) | | Metric | Spearman Correlation (ρs) | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 30 of 30 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Spearman Correlation (ρs)](): Higher is better; unit: SCALAR. ## Recorded results | Method | Spearman Correlation (ρs) | Result date | Report paper | | --- | --- | --- | --- | | CLARE | 0.918 | Not recorded | 69c0c9020636e5bf4afc3b5a | | CLARE | 0.917 | Not recorded | 69c0c9020636e5bf4afc3b5a | | CLARE | 0.915 | Not recorded | 69c0c9020636e5bf4afc3b5a | | CLARE | 0.913 | Not recorded | 69c0c9020636e5bf4afc3b5a | | CLARE | 0.911 | Not recorded | 69c0c9020636e5bf4afc3b5a | | CLARE | 0.909 | Not recorded | 69c0c9020636e5bf4afc3b5a | | CLARE | 0.878 | Not recorded | 69c0c9020636e5bf4afc3b5a | | CLARE | 0.876 | Not recorded | 69c0c9020636e5bf4afc3b5a | | CLARE | 0.875 | Not recorded | 69c0c9020636e5bf4afc3b5a | | CLARE | 0.871 | Not recorded | 69c0c9020636e5bf4afc3b5a | | CLARE | 0.861 | Not recorded | 69c0c9020636e5bf4afc3b5a | | CLARE | 0.857 | Not recorded | 69c0c9020636e5bf4afc3b5a | | CLARE | 0.854 | Not recorded | 69c0c9020636e5bf4afc3b5a | | CLARE | 0.848 | Not recorded | 69c0c9020636e5bf4afc3b5a | | CLARE | 0.801 | Not recorded | 69c0c9020636e5bf4afc3b5a | | GradSim | 0.691 | Not recorded | 69c0c9020636e5bf4afc3b5a | | GradSim | 0.677 | Not recorded | 69c0c9020636e5bf4afc3b5a | | GradSim | 0.663 | Not recorded | 69c0c9020636e5bf4afc3b5a | | GradSim | 0.646 | Not recorded | 69c0c9020636e5bf4afc3b5a | | GradSim | 0.634 | Not recorded | 69c0c9020636e5bf4afc3b5a | | GradSim | 0.561 | Not recorded | 69c0c9020636e5bf4afc3b5a | | GradSim | 0.558 | Not recorded | 69c0c9020636e5bf4afc3b5a | | GradSim | 0.554 | Not recorded | 69c0c9020636e5bf4afc3b5a | | GradSim | 0.548 | Not recorded | 69c0c9020636e5bf4afc3b5a | | GradSim | 0.543 | Not recorded | 69c0c9020636e5bf4afc3b5a | | GradSim | 0.468 | Not recorded | 69c0c9020636e5bf4afc3b5a | | GradSim | 0.46 | Not recorded | 69c0c9020636e5bf4afc3b5a | | GradSim | 0.442 | Not recorded | 69c0c9020636e5bf4afc3b5a | | GradSim | 0.442 | Not recorded | 69c0c9020636e5bf4afc3b5a | | GradSim | 0.389 | Not recorded | 69c0c9020636e5bf4afc3b5a | ### CLARE Spearman Correlation (ρs): 0.918. Result ID: 69c11e2547f34ce924a96dd3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT2-XL","Editing Technique":"PRUNE"}. This record does not assert independent reproduction. ### CLARE Spearman Correlation (ρs): 0.917. Result ID: 69c11e2547f34ce924a96dc9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-J","Editing Technique":"PRUNE"}. This record does not assert independent reproduction. ### CLARE Spearman Correlation (ρs): 0.915. Result ID: 69c11e2547f34ce924a96dcd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT2-XL","Editing Technique":"MEMIT"}. This record does not assert independent reproduction. ### CLARE Spearman Correlation (ρs): 0.913. Result ID: 69c11e2547f34ce924a96dc3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-J","Editing Technique":"MEMIT"}. This record does not assert independent reproduction. ### CLARE Spearman Correlation (ρs): 0.911. Result ID: 69c11e2547f34ce924a96dcf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT2-XL","Editing Technique":"RECT"}. This record does not assert independent reproduction. ### CLARE Spearman Correlation (ρs): 0.909. Result ID: 69c11e2547f34ce924a96dd1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT2-XL","Editing Technique":"ROME"}. This record does not assert independent reproduction. ### CLARE Spearman Correlation (ρs): 0.878. Result ID: 69c11e2547f34ce924a96dc5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-J","Editing Technique":"RECT"}. This record does not assert independent reproduction. ### CLARE Spearman Correlation (ρs): 0.876. Result ID: 69c11e2547f34ce924a96ddf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama3","Editing Technique":"AlphaEdit"}. This record does not assert independent reproduction. ### CLARE Spearman Correlation (ρs): 0.875. Result ID: 69c11e2547f34ce924a96dcb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-J","Editing Technique":"AlphaEdit"}. This record does not assert independent reproduction. ### CLARE Spearman Correlation (ρs): 0.871. Result ID: 69c11e2547f34ce924a96ddd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama3","Editing Technique":"PRUNE"}. This record does not assert independent reproduction. ### CLARE Spearman Correlation (ρs): 0.861. Result ID: 69c11e2547f34ce924a96dd9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama3","Editing Technique":"RECT"}. This record does not assert independent reproduction. ### CLARE Spearman Correlation (ρs): 0.857. Result ID: 69c11e2547f34ce924a96dc7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-J","Editing Technique":"ROME"}. This record does not assert independent reproduction. ### CLARE Spearman Correlation (ρs): 0.854. Result ID: 69c11e2547f34ce924a96dd5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT2-XL","Editing Technique":"AlphaEdit"}. This record does not assert independent reproduction. ### CLARE Spearman Correlation (ρs): 0.848. Result ID: 69c11e2547f34ce924a96dd7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama3","Editing Technique":"MEMIT"}. This record does not assert independent reproduction. ### CLARE Spearman Correlation (ρs): 0.801. Result ID: 69c11e2547f34ce924a96ddb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama3","Editing Technique":"ROME"}. This record does not assert independent reproduction. ### GradSim Spearman Correlation (ρs): 0.691. Result ID: 69c11e2547f34ce924a96dc8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-J","Editing Technique":"PRUNE"}. This record does not assert independent reproduction. ### GradSim Spearman Correlation (ρs): 0.677. Result ID: 69c11e2547f34ce924a96dc2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-J","Editing Technique":"MEMIT"}. This record does not assert independent reproduction. ### GradSim Spearman Correlation (ρs): 0.663. Result ID: 69c11e2547f34ce924a96dca. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-J","Editing Technique":"AlphaEdit"}. This record does not assert independent reproduction. ### GradSim Spearman Correlation (ρs): 0.646. Result ID: 69c11e2547f34ce924a96dc4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-J","Editing Technique":"RECT"}. This record does not assert independent reproduction. ### GradSim Spearman Correlation (ρs): 0.634. Result ID: 69c11e2547f34ce924a96dc6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-J","Editing Technique":"ROME"}. This record does not assert independent reproduction. ### GradSim Spearman Correlation (ρs): 0.561. Result ID: 69c11e2547f34ce924a96dd2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT2-XL","Editing Technique":"PRUNE"}. This record does not assert independent reproduction. ### GradSim Spearman Correlation (ρs): 0.558. Result ID: 69c11e2547f34ce924a96dd0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT2-XL","Editing Technique":"ROME"}. This record does not assert independent reproduction. ### GradSim Spearman Correlation (ρs): 0.554. Result ID: 69c11e2547f34ce924a96dd4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT2-XL","Editing Technique":"AlphaEdit"}. This record does not assert independent reproduction. ### GradSim Spearman Correlation (ρs): 0.548. Result ID: 69c11e2547f34ce924a96dce. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT2-XL","Editing Technique":"RECT"}. This record does not assert independent reproduction. ### GradSim Spearman Correlation (ρs): 0.543. Result ID: 69c11e2547f34ce924a96dcc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT2-XL","Editing Technique":"MEMIT"}. This record does not assert independent reproduction. ### GradSim Spearman Correlation (ρs): 0.468. Result ID: 69c11e2547f34ce924a96dd8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama3","Editing Technique":"RECT"}. This record does not assert independent reproduction. ### GradSim Spearman Correlation (ρs): 0.46. Result ID: 69c11e2547f34ce924a96ddc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama3","Editing Technique":"PRUNE"}. This record does not assert independent reproduction. ### GradSim Spearman Correlation (ρs): 0.442. Result ID: 69c11e2547f34ce924a96dd6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama3","Editing Technique":"MEMIT"}. This record does not assert independent reproduction. ### GradSim Spearman Correlation (ρs): 0.442. Result ID: 69c11e2547f34ce924a96dde. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama3","Editing Technique":"AlphaEdit"}. This record does not assert independent reproduction. ### GradSim Spearman Correlation (ρs): 0.389. Result ID: 69c11e2547f34ce924a96dda. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama3","Editing Technique":"ROME"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.