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

    Neuronpedia

    INDEX
    Explanations

    visible

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     bells
    -0.08
     обратиться
    -0.08
     обращаться
    -0.08
    pager
    -0.08
     reward
    -0.08
     voting
    -0.08
    909
    -0.07
    ольз
    -0.07
     adjud
    -0.07
     scoring
    -0.07
    POSITIVE LOGITS
    空气
    0.10
    _air
    0.09
     aerosol
    0.09
     condens
    0.09
     Air
    0.09
     PROFILE
    0.08
     हवा
    0.08
     gchar
    0.08
     aeros
    0.08
    -air
    0.08
    Act Density 0.008%

    No Known Activations