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

    Neuronpedia

    INDEX
    Explanations

    ends in eaten

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    américa
    0.40
     memcpy
    0.39
     donating
    0.39
    養
    0.38
     Acquisition
    0.38
     Sherman
    0.38
    elsius
    0.37
     compartments
    0.37
     Introduce
    0.37
    parsing
    0.37
    POSITIVE LOGITS
    oken
    0.77
     eaten
    0.75
    en
    0.75
    taken
    0.71
    ken
    0.71
     beaten
    0.69
     ridden
    0.69
    KEN
    0.67
     taken
    0.67
    ollen
    0.66
    Act Density 0.014%

    No Known Activations