© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B
    5. 31-GEMMASCOPE-2-RES-262K
    7. 2073
    Prev
    Next
    INDEX
    Explanations

    gender and nationality

    np_acts-logits-general · gemini-2.5-flash-lite

    without explicit content

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    越来越多
    0.42
    磷
    0.42
     बढ़
    0.41
    
    0.41
    數
    0.41
    ጮ
    0.41
    량
    0.40
     করেছি
    0.40
     уте
    0.40
    츰
    0.40
    POSITIVE LOGITS
     will
    0.55
     being
    0.54
     gender
    0.51
     nebude
    0.50
     they
    0.49
     Gender
    0.49
     assumes
    0.48
     nationality
    0.48
     assumed
    0.46
     genders
    0.46
    Activations Density 0.002%

    No Known Activations