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

    Neuronpedia

    INDEX
    Explanations

    Britain

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     themſelves
    -0.87
     ſtate
    -0.87
    complexContent
    -0.78
    hematical
    -0.77
     punishment
    -0.77
    ugeot
    -0.77
     poverty
    -0.76
    ſelves
    -0.76
    Дереккөздер
    -0.76
     purpoſe
    -0.76
    POSITIVE LOGITS
    s
    0.73
    v
    0.59
    f
    0.56
    if
    0.54
    w
    0.54
    odo
    0.53
     in
    0.52
    r
    0.52
    ra
    0.51
    h
    0.51
    Act Density 1.314%

    No Known Activations