# Circuit Recovery on Aggregate IOI, MCQA, Arithmetic, ARC

Source: [Circuit Recovery on Aggregate IOI, MCQA, Arithmetic, ARC](<https://www.sota2.com/research/sota/circuit-recovery-on-aggregate-ioi-mcqa-arithmetic-arc>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-28T10:11:43.638Z

Evaluation of model performance on the aggregated Circuit Recovery benchmark combining results from IOI, MCQA, Arithmetic, and ARC datasets, measured by the CPR faithfulness score.

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 Recovery |
| Dataset (including any stated split) | Aggregate IOI, MCQA, Arithmetic, ARC |
| Metric | CPR Faithfulness Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 22 of 22 |
| 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 Faithfulness Score](<https://www.sota2.com/research/sota/circuit-recovery-on-aggregate-ioi-mcqa-arithmetic-arc.md?metric=CPR%20Faithfulness%20Score>): Higher is better; unit: PERCENT.

- [IOI Faithfulness Score](<https://www.sota2.com/research/sota/circuit-recovery-on-aggregate-ioi-mcqa-arithmetic-arc.md?metric=IOI%20Faithfulness%20Score>): Higher is better; unit: PERCENT.

- [MCQA Faithfulness Score](<https://www.sota2.com/research/sota/circuit-recovery-on-aggregate-ioi-mcqa-arithmetic-arc.md?metric=MCQA%20Faithfulness%20Score>): Higher is better; unit: PERCENT.

- [Arithmetic+ Faithfulness Score](<https://www.sota2.com/research/sota/circuit-recovery-on-aggregate-ioi-mcqa-arithmetic-arc.md?metric=Arithmetic%2B%20Faithfulness%20Score>): Higher is better; unit: PERCENT.

- [Arithmetic- Faithfulness Score](<https://www.sota2.com/research/sota/circuit-recovery-on-aggregate-ioi-mcqa-arithmetic-arc.md?metric=Arithmetic-%20Faithfulness%20Score>): Higher is better; unit: PERCENT.

- [ARC-E Faithfulness Score](<https://www.sota2.com/research/sota/circuit-recovery-on-aggregate-ioi-mcqa-arithmetic-arc.md?metric=ARC-E%20Faithfulness%20Score>): Higher is better; unit: PERCENT.

- [ARC-C Faithfulness Score](<https://www.sota2.com/research/sota/circuit-recovery-on-aggregate-ioi-mcqa-arithmetic-arc.md?metric=ARC-C%20Faithfulness%20Score>): Higher is better; unit: PERCENT.

- [Average Faithfulness Score](<https://www.sota2.com/research/sota/circuit-recovery-on-aggregate-ioi-mcqa-arithmetic-arc.md?metric=Average%20Faithfulness%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | CPR Faithfulness Score | Result date | Report paper |
| --- | --- | --- | --- |
| NAP-IG-Inputs | 87 | Not recorded | 69f03f7555a3c32af89fa4df |
| NAP-IG-Activations | 68 | Not recorded | 69f03f7555a3c32af89fa4df |
| NAP | 66 | Not recorded | 69f03f7555a3c32af89fa4df |
| NAP-IG-Inputs | 31 | Not recorded | 69f03f7555a3c32af89fa4df |
| NAP | 31 | Not recorded | 69f03f7555a3c32af89fa4df |
| NAP-IG-Activations | 30 | Not recorded | 69f03f7555a3c32af89fa4df |
| NAP-IG-Inputs | 29 | Not recorded | 69f03f7555a3c32af89fa4df |
| NAP-IG-Inputs | 29 | Not recorded | 69f03f7555a3c32af89fa4df |
| NAP-IG-Activations | 28 | Not recorded | 69f03f7555a3c32af89fa4df |
| NAP | 27 | Not recorded | 69f03f7555a3c32af89fa4df |
| NAP | 27 | Not recorded | 69f03f7555a3c32af89fa4df |
| NAP-IG-Activations | 27 | Not recorded | 69f03f7555a3c32af89fa4df |
| Baseline | 26 | Not recorded | 69f03f7555a3c32af89fa4df |

### NAP-IG-Inputs

CPR Faithfulness Score: 87. Result ID: 69f087cc47f34ce924d70ed7.

[Reporting paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

Recorded evaluation configuration: {"Method Category":"NAP-IG-Inputs","Transfer Setting":"Gold (Upper Bound)"}. This record does not assert independent reproduction.

### NAP-IG-Activations

CPR Faithfulness Score: 68. Result ID: 69f087cc47f34ce924d70edf.

[Reporting paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

Recorded evaluation configuration: {"Method Category":"NAP-IG-Activations","Transfer Setting":"Gold (Upper Bound)"}. This record does not assert independent reproduction.

### NAP

CPR Faithfulness Score: 66. Result ID: 69f087cc47f34ce924d70edb.

[Reporting paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

Recorded evaluation configuration: {"Method Category":"NAP","Transfer Setting":"Gold (Upper Bound)"}. This record does not assert independent reproduction.

### NAP-IG-Inputs

CPR Faithfulness Score: 31. Result ID: 69f087cc47f34ce924d70eda.

[Reporting paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

Recorded evaluation configuration: {"Method Category":"NAP-IG-Inputs","Transfer Setting":"DFA (Far/Zero-Shot)"}. This record does not assert independent reproduction.

### NAP

CPR Faithfulness Score: 31. Result ID: 69f087cc47f34ce924d70ede.

[Reporting paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

Recorded evaluation configuration: {"Method Category":"NAP","Transfer Setting":"DFA (Far/Zero-Shot)"}. This record does not assert independent reproduction.

### NAP-IG-Activations

CPR Faithfulness Score: 30. Result ID: 69f087cc47f34ce924d70ee2.

[Reporting paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

Recorded evaluation configuration: {"Method Category":"NAP-IG-Activations","Transfer Setting":"DFA (Far/Zero-Shot)"}. This record does not assert independent reproduction.

### NAP-IG-Inputs

CPR Faithfulness Score: 29. Result ID: 69f087cc47f34ce924d70ed8.

[Reporting paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

Recorded evaluation configuration: {"Method Category":"NAP-IG-Inputs","Transfer Setting":"DFA (In-Distribution)"}. This record does not assert independent reproduction.

### NAP-IG-Inputs

CPR Faithfulness Score: 29. Result ID: 69f087cc47f34ce924d70ed9.

[Reporting paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

Recorded evaluation configuration: {"Method Category":"NAP-IG-Inputs","Transfer Setting":"DFA (Near-Distribution)"}. This record does not assert independent reproduction.

### NAP-IG-Activations

CPR Faithfulness Score: 28. Result ID: 69f087cc47f34ce924d70ee0.

[Reporting paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

Recorded evaluation configuration: {"Method Category":"NAP-IG-Activations","Transfer Setting":"DFA (In-Distribution)"}. This record does not assert independent reproduction.

### NAP

CPR Faithfulness Score: 27. Result ID: 69f087cc47f34ce924d70edc.

[Reporting paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

Recorded evaluation configuration: {"Method Category":"NAP","Transfer Setting":"DFA (In-Distribution)"}. This record does not assert independent reproduction.

### NAP

CPR Faithfulness Score: 27. Result ID: 69f087cc47f34ce924d70edd.

[Reporting paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

Recorded evaluation configuration: {"Method Category":"NAP","Transfer Setting":"DFA (Near-Distribution)"}. This record does not assert independent reproduction.

### NAP-IG-Activations

CPR Faithfulness Score: 27. Result ID: 69f087cc47f34ce924d70ee1.

[Reporting paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

Recorded evaluation configuration: {"Method Category":"NAP-IG-Activations","Transfer Setting":"DFA (Near-Distribution)"}. This record does not assert independent reproduction.

### Baseline

CPR Faithfulness Score: 26. Result ID: 69f087cc47f34ce924d70ed6.

[Reporting paper](<https://www.sota2.com/research/paper/69f03f7555a3c32af89fa4df.md>)

Recorded evaluation configuration: {"Method Category":"Baseline","Transfer Setting":"Random Alignment"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.