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

    Neuronpedia

    INDEX
    Explanations

    understand, extract, influence

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     but
    0.56
     in
    0.53
     with
    0.51
     from
    0.50
     aspirated
    0.48
     for
    0.48
     of
    0.47
     between
    0.45
     here
    0.45
     on
    0.44
    POSITIVE LOGITS
    Loksatta
    0.55
    UGH
    0.53
     забезпе
    0.53
     อำ
    0.51
    Maximize
    0.49
    ERRY
    0.49
     zapewn
    0.49
    显著
    0.48
    营造
    0.47
    Funcion
    0.47
    Act Density 0.018%

    No Known Activations