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

    Neuronpedia

    INDEX
    Explanations

    think of it like this

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     verge
    0.39
    mios
    0.38
     tech
    0.38
     durations
    0.38
     aiding
    0.37
     What
    0.37
     indeed
    0.36
    ρεία
    0.36
     Verge
    0.36
     activate
    0.36
    POSITIVE LOGITS
     자동차
    0.82
    汽车
    0.77
    汽車
    0.71
    자동차
    0.71
     analogy
    0.64
     samoch
    0.64
    automobile
    0.64
     lego
    0.63
     automobiles
    0.62
     carpenters
    0.61
    Act Density 0.100%

    No Known Activations