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

    Neuronpedia

    INDEX
    Explanations

    off topic, off load, off grid

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    fool
    0.42
    かわいい
    0.42
     sensibly
    0.41
    ^{*}}
    0.40
    Produ
    0.39
    bai
    0.39
    enhance
    0.39
    both
    0.39
    abortion
    0.38
    bacteria
    0.38
    POSITIVE LOGITS
     Off
    0.83
    izielle
    0.70
    icial
    0.67
     off
    0.66
    Off
    0.65
     course
    0.63
    off
    0.61
     beaten
    0.61
    ensive
    0.60
     topic
    0.60
    Act Density 0.024%

    No Known Activations