# Circuit Localization on Indirect Object Identification (IOI) 1.0 (test) Source: [Circuit Localization on Indirect Object Identification (IOI) 1.0 (test)]() Collection index: [llms.txt]() Data updated: 2026-05-12T11:18:36.187Z Evaluation of model performance on the test set of the Indirect Object Identification (IOI) 1.0 dataset for the Circuit Localization task, measured by CPR. 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 | Circuit Localization | | Dataset (including any stated split) | Indirect Object Identification (IOI) 1.0 (test) | | Metric | CPR | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 28 of 28 | | 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 - [CPR](): Higher is better; unit: SCALAR. ## Recorded results | Method | CPR | Result date | Report paper | | --- | --- | --- | --- | | DCD-kmeans-PCA | 1.068 | Not recorded | 6a02b45cdbc836b87d8249e8 | | DCD-divisive-SVD | 1.054 | Not recorded | 6a02b45cdbc836b87d8249e8 | | DCD-kmeans-SVD | 1.045 | Not recorded | 6a02b45cdbc836b87d8249e8 | | DCD-kmeans-PCA | 1.045 | Not recorded | 6a02b45cdbc836b87d8249e8 | | K-Random | 1.038 | Not recorded | 6a02b45cdbc836b87d8249e8 | | DCD-agglom.-SVD | 1.038 | Not recorded | 6a02b45cdbc836b87d8249e8 | | K-Random | 1.037 | Not recorded | 6a02b45cdbc836b87d8249e8 | | DCD-kmeans-SVD | 1.036 | Not recorded | 6a02b45cdbc836b87d8249e8 | | DCD-agglom.-SVD | 1.024 | Not recorded | 6a02b45cdbc836b87d8249e8 | | DCD-divisive-SVD | 1.024 | Not recorded | 6a02b45cdbc836b87d8249e8 | | K-Random | 1.021 | Not recorded | 6a02b45cdbc836b87d8249e8 | | K-representative | 1.015 | Not recorded | 6a02b45cdbc836b87d8249e8 | | DCD-kmeans-PCA | 0.998 | Not recorded | 6a02b45cdbc836b87d8249e8 | | EAP-IG | 0.997 | Not recorded | 6a02b45cdbc836b87d8249e8 | | K-representative | 0.997 | Not recorded | 6a02b45cdbc836b87d8249e8 | | DCD-kmeans-SVD | 0.994 | Not recorded | 6a02b45cdbc836b87d8249e8 | | DCD-divisive-SVD | 0.99 | Not recorded | 6a02b45cdbc836b87d8249e8 | | EAP | 0.984 | Not recorded | 6a02b45cdbc836b87d8249e8 | | EAP-IG | 0.98 | Not recorded | 6a02b45cdbc836b87d8249e8 | | DCD-agglom.-SVD | 0.954 | Not recorded | 6a02b45cdbc836b87d8249e8 | | K-representative | 0.949 | Not recorded | 6a02b45cdbc836b87d8249e8 | | EAP | 0.936 | Not recorded | 6a02b45cdbc836b87d8249e8 | | EAP | 0.931 | Not recorded | 6a02b45cdbc836b87d8249e8 | | EAP-IG | 0.873 | Not recorded | 6a02b45cdbc836b87d8249e8 | | E-Act | 0.87 | Not recorded | 6a02b45cdbc836b87d8249e8 | | Random | 0.25 | Not recorded | 6a02b45cdbc836b87d8249e8 | | Random | 0.245 | Not recorded | 6a02b45cdbc836b87d8249e8 | | Random | 0.153 | Not recorded | 6a02b45cdbc836b87d8249e8 | ### DCD-kmeans-PCA CPR: 1.068. Result ID: 6a030c7d47f34ce924e6cbab. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen"}. This record does not assert independent reproduction. ### DCD-divisive-SVD CPR: 1.054. Result ID: 6a030c7d47f34ce924e6cba5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen"}. This record does not assert independent reproduction. ### DCD-kmeans-SVD CPR: 1.045. Result ID: 6a030c7d47f34ce924e6cba8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen"}. This record does not assert independent reproduction. ### DCD-kmeans-PCA CPR: 1.045. Result ID: 6a030c7d47f34ce924e6cbaa. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama"}. This record does not assert independent reproduction. ### K-Random CPR: 1.038. Result ID: 6a030c7d47f34ce924e6cb9c. [Reporting paper]() Recorded evaluation configuration: {"Model":"Qwen"}. This record does not assert independent reproduction. ### DCD-agglom.-SVD CPR: 1.038. Result ID: 6a030c7d47f34ce924e6cba2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen"}. This record does not assert independent reproduction. ### K-Random CPR: 1.037. Result ID: 6a030c7d47f34ce924e6cb9b. [Reporting paper]() Recorded evaluation configuration: {"Model":"Llama"}. This record does not assert independent reproduction. ### DCD-kmeans-SVD CPR: 1.036. Result ID: 6a030c7d47f34ce924e6cba7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama"}. This record does not assert independent reproduction. ### DCD-agglom.-SVD CPR: 1.024. Result ID: 6a030c7d47f34ce924e6cba1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama"}. This record does not assert independent reproduction. ### DCD-divisive-SVD CPR: 1.024. Result ID: 6a030c7d47f34ce924e6cba4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama"}. This record does not assert independent reproduction. ### K-Random CPR: 1.021. Result ID: 6a030c7d47f34ce924e6cb9a. [Reporting paper]() Recorded evaluation configuration: {"Model":"GPT2"}. This record does not assert independent reproduction. ### K-representative CPR: 1.015. Result ID: 6a030c7d47f34ce924e6cb9e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama"}. This record does not assert independent reproduction. ### DCD-kmeans-PCA CPR: 0.998. Result ID: 6a030c7d47f34ce924e6cba9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT2"}. This record does not assert independent reproduction. ### EAP-IG CPR: 0.997. Result ID: 6a030c7d47f34ce924e6cb91. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama"}. This record does not assert independent reproduction. ### K-representative CPR: 0.997. Result ID: 6a030c7d47f34ce924e6cb9f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen"}. This record does not assert independent reproduction. ### DCD-kmeans-SVD CPR: 0.994. Result ID: 6a030c7d47f34ce924e6cba6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT2"}. This record does not assert independent reproduction. ### DCD-divisive-SVD CPR: 0.99. Result ID: 6a030c7d47f34ce924e6cba3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT2"}. This record does not assert independent reproduction. ### EAP CPR: 0.984. Result ID: 6a030c7d47f34ce924e6cb94. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama"}. This record does not assert independent reproduction. ### EAP-IG CPR: 0.98. Result ID: 6a030c7d47f34ce924e6cb92. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen"}. This record does not assert independent reproduction. ### DCD-agglom.-SVD CPR: 0.954. Result ID: 6a030c7d47f34ce924e6cba0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT2"}. This record does not assert independent reproduction. ### K-representative CPR: 0.949. Result ID: 6a030c7d47f34ce924e6cb9d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT2"}. This record does not assert independent reproduction. ### EAP CPR: 0.936. Result ID: 6a030c7d47f34ce924e6cb95. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen"}. This record does not assert independent reproduction. ### EAP CPR: 0.931. Result ID: 6a030c7d47f34ce924e6cb93. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT2"}. This record does not assert independent reproduction. ### EAP-IG CPR: 0.873. Result ID: 6a030c7d47f34ce924e6cb90. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT2"}. This record does not assert independent reproduction. ### E-Act CPR: 0.87. Result ID: 6a030c7d47f34ce924e6cb96. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"GPT2"}. This record does not assert independent reproduction. ### Random CPR: 0.25. Result ID: 6a030c7d47f34ce924e6cb98. [Reporting paper]() Recorded evaluation configuration: {"Model":"Llama"}. This record does not assert independent reproduction. ### Random CPR: 0.245. Result ID: 6a030c7d47f34ce924e6cb99. [Reporting paper]() Recorded evaluation configuration: {"Model":"Qwen"}. This record does not assert independent reproduction. ### Random CPR: 0.153. Result ID: 6a030c7d47f34ce924e6cb97. [Reporting paper]() Recorded evaluation configuration: {"Model":"GPT2"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.