# Gender Neuron Masking on InclusiveGender Feminine Source: [Gender Neuron Masking on InclusiveGender Feminine]() Collection index: [llms.txt]() Data updated: 2026-06-01T08:04:25.651Z Evaluation of model performance on masking gendered neurons using the InclusiveGender Feminine dataset, focusing on term counts and mention ratios across male (M), female (F), and neutral (N) categories. 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 Feminine | | Metric | Count of Gendered Terms per Response (M) | | Unit | SCALAR | | Direction | Higher 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 - [Count of Gendered Terms per Response (M)](): Higher is better; unit: SCALAR. - [Count of Gendered Terms per Response (F)](): Higher is better; unit: SCALAR. - [Count of Gendered Terms per Response (N)](): Higher is better; unit: SCALAR. - [Mention Ratio (M)](): Higher is better; unit: PERCENT. - [Mention Ratio (F)](): Higher is better; unit: PERCENT. - [Mention Ratio (N)](): Higher is better; unit: PERCENT. - [Delta Ratio (M)](): Lower is better; unit: PERCENT. - [Delta Ratio (F)](): Lower is better; unit: PERCENT. - [Delta Ratio (N)](): Lower is better; unit: PERCENT. ## Recorded results | Method | Count of Gendered Terms per Response (M) | Result date | Report paper | | --- | --- | --- | --- | | sNeuron-TST | 0.8 | Not recorded | 6a1d1213dbd7150506f83fef | | Baseline | 0.38 | Not recorded | 6a1d1213dbd7150506f83fef | | Ours | 0.33 | Not recorded | 6a1d1213dbd7150506f83fef | | LAPE | 0.28 | Not recorded | 6a1d1213dbd7150506f83fef | ### sNeuron-TST Count of Gendered Terms per Response (M): 0.8. Result ID: 6a1d3c4cfbc373592c163a1f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama 3.1"}. This record does not assert independent reproduction. ### Baseline Count of Gendered Terms per Response (M): 0.38. Result ID: 6a1d3c4cfbc373592c163a1e. [Reporting paper]() Recorded evaluation configuration: {"Base Model":"Llama 3.1"}. This record does not assert independent reproduction. ### Ours Count of Gendered Terms per Response (M): 0.33. Result ID: 6a1d3c4cfbc373592c163a21. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Llama 3.1"}. This record does not assert independent reproduction. ### LAPE Count of Gendered Terms per Response (M): 0.28. Result ID: 6a1d3c4cfbc373592c163a20. [Reporting paper]() [Method origin paper]() [Implementation]() 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.