# Gender Neuron Masking on InclusiveGender Masculine

Source: [Gender Neuron Masking on InclusiveGender Masculine](<https://www.sota2.com/research/sota/gender-neuron-masking-on-inclusivegender-masculine>)

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

Data updated: 2026-06-01T08:04:25.678Z

Evaluation of the Gender Neuron Masking task using the InclusiveGender Masculine dataset, focusing on the distribution and delta ratios of gendered terms (M, F, N).

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 | Gender Neuron Masking |
| Dataset (including any stated split) | InclusiveGender Masculine |
| Metric | Avg. Gendered Terms per Response (M) |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 4 of 4 |
| 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

- [Avg. Gendered Terms per Response (M)](<https://www.sota2.com/research/sota/gender-neuron-masking-on-inclusivegender-masculine.md?metric=Avg.%20Gendered%20Terms%20per%20Response%20(M)>): Lower is better; unit: SCALAR.

- [Avg. Gendered Terms per Response (F)](<https://www.sota2.com/research/sota/gender-neuron-masking-on-inclusivegender-masculine.md?metric=Avg.%20Gendered%20Terms%20per%20Response%20(F)>): Lower is better; unit: SCALAR.

- [Avg. Gendered Terms per Response (N)](<https://www.sota2.com/research/sota/gender-neuron-masking-on-inclusivegender-masculine.md?metric=Avg.%20Gendered%20Terms%20per%20Response%20(N)>): Lower is better; unit: SCALAR.

- [Mention Ratio (%) (M)](<https://www.sota2.com/research/sota/gender-neuron-masking-on-inclusivegender-masculine.md?metric=Mention%20Ratio%20(%25)%20(M)>): Lower is better; unit: PERCENT.

- [Mention Ratio (%) (F)](<https://www.sota2.com/research/sota/gender-neuron-masking-on-inclusivegender-masculine.md?metric=Mention%20Ratio%20(%25)%20(F)>): Lower is better; unit: PERCENT.

- [Mention Ratio (%) (N)](<https://www.sota2.com/research/sota/gender-neuron-masking-on-inclusivegender-masculine.md?metric=Mention%20Ratio%20(%25)%20(N)>): Lower is better; unit: PERCENT.

- [Delta Ratio (%) (M)](<https://www.sota2.com/research/sota/gender-neuron-masking-on-inclusivegender-masculine.md?metric=Delta%20Ratio%20(%25)%20(M)>): Lower is better; unit: PERCENT.

- [Delta Ratio (%) (F)](<https://www.sota2.com/research/sota/gender-neuron-masking-on-inclusivegender-masculine.md?metric=Delta%20Ratio%20(%25)%20(F)>): Higher is better; unit: PERCENT.

- [Delta Ratio (%) (N)](<https://www.sota2.com/research/sota/gender-neuron-masking-on-inclusivegender-masculine.md?metric=Delta%20Ratio%20(%25)%20(N)>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Avg. Gendered Terms per Response (M) | Result date | Report paper |
| --- | --- | --- | --- |
| LAPE | 1.39 | Not recorded | 6a1d1213dbd7150506f83fef |
| sNeuron-TST | 1.56 | Not recorded | 6a1d1213dbd7150506f83fef |
| Baseline | 1.58 | Not recorded | 6a1d1213dbd7150506f83fef |
| Ours | 1.6 | Not recorded | 6a1d1213dbd7150506f83fef |

### LAPE

Avg. Gendered Terms per Response (M): 1.39. Result ID: 6a1d3c4cfbc373592c163a24.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d1213dbd7150506f83fef.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738b1e209131cea5d1c2.md>)

[Implementation](<https://github.com/RUCAIBox/Language-Specific-Neurons>)

Recorded evaluation configuration: {"Base Model":"Llama 3.1"}. This record does not assert independent reproduction.

### sNeuron-TST

Avg. Gendered Terms per Response (M): 1.56. Result ID: 6a1d3c4cfbc373592c163a23.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d1213dbd7150506f83fef.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75081e209131cea65682.md>)

Recorded evaluation configuration: {"Base Model":"Llama 3.1"}. This record does not assert independent reproduction.

### Baseline

Avg. Gendered Terms per Response (M): 1.58. Result ID: 6a1d3c4cfbc373592c163a22.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d1213dbd7150506f83fef.md>)

Recorded evaluation configuration: {"Base Model":"Llama 3.1"}. This record does not assert independent reproduction.

### Ours

Avg. Gendered Terms per Response (M): 1.6. Result ID: 6a1d3c4cfbc373592c163a25.

[Reporting paper](<https://www.sota2.com/research/paper/6a1d1213dbd7150506f83fef.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a1d1213dbd7150506f83fef.md>)

Recorded evaluation configuration: {"Base Model":"Llama 3.1"}. This record does not assert independent reproduction.

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