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

    Neuronpedia

    INDEX
    Explanations

    tickets escalated, changed, or lost

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     gestalten
    0.79
    defined
    0.78
    informed
    0.78
     Defined
    0.77
    Defined
    0.76
     Informed
    0.74
     informed
    0.71
     établie
    0.68
     заданной
    0.67
     defined
    0.66
    POSITIVE LOGITS
     lost
    1.59
     transferred
    1.48
     removed
    1.38
     swallowed
    1.34
     snatched
    1.33
     absorbed
    1.32
     moved
    1.31
     transported
    1.29
     stolen
    1.25
     sent
    1.24
    Act Density 0.365%

    No Known Activations