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

    Neuronpedia

    INDEX
    Explanations

    and, so

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Such
    -0.72
     Then
    -0.72
     Again
    -0.69
    .
    -0.64
     Many
    -0.64
     They
    -0.62
     Much
    -0.61
     Its
    -0.61
     Rather
    -0.60
    Then
    -0.59
    POSITIVE LOGITS
     the
    0.92
    <bos>
    0.75
     you
    0.73
     it
    0.71
     that
    0.70
     Савезне
    0.67
     we
    0.64
     if
    0.63
     when
    0.63
     initComponents
    0.63
    Act Density 0.048%

    No Known Activations