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

    Neuronpedia

    INDEX
    Explanations

    reading or opening with `as`

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     been
    -1.41
     since
    -1.40
     was
    -1.33
     должна
    -1.27
    3
    -1.25
     already
    -1.23
    1
    -1.23
    2
    -1.21
     should
    -1.21
    begin
    -1.21
    POSITIVE LOGITS
    itados
    1.53
    isième
    1.39
    plevel
    1.36
     {
    1.35
    崢
    1.33
    辞职
    1.31
    ències
    1.25
    abile
    1.25
    dań
    1.23
    inės
    1.22
    Act Density 0.040%

    No Known Activations