Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    orth

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     contracting
    -0.87
    orth
    -0.78
     Po
    -0.77
    Po
    -0.65
     missing
    -0.63
    es
    -0.62
     contracted
    -0.62
     Contracting
    -0.60
    k
    -0.60
    or
    -0.56
    POSITIVE LOGITS
    rrggbb
    0.82
    Rüyada
    0.71
     мәкал
    0.70
    帖最后由
    0.70
     propOrder
    0.69
    ientôt
    0.68
    umptive
    0.67
    Tikang
    0.66
    IVEREF
    0.66
     GenerationType
    0.63
    Act Density 0.420%

    No Known Activations