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

    Neuronpedia

    INDEX
    Explanations

    heat temperature cooling

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     prič
    2.07
    ு
    1.98
    uintes
    1.94
    ernsey
    1.93
     patrón
    1.88
    眙
    1.87
    erode
    1.85
    必然
    1.85
    椏
    1.82
    anediyl
    1.82
    POSITIVE LOGITS
    炉
    1.86
     dhamme
    1.82
    忱
    1.75
    ey
    1.74
     blankets
    1.72
    ทรัพย์
    1.70
    溫度
    1.69
     हस्ता
    1.67
     furnaces
    1.66
    爐
    1.65
    Act Density 0.667%

    No Known Activations