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

    Neuronpedia

    INDEX
    Explanations

    describing writing qualities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    of
    0.59
    使用
    0.55
            
    0.50
     of
    0.47
     $\
    0.46
    ahl
    0.44
    ("
    0.43
    using
    0.43
    ^{
    0.42
    >
    0.41
    POSITIVE LOGITS
    ون
    0.54
    ور
    0.49
     vigil
    0.44
     paradise
    0.44
     счастли
    0.43
     haci
    0.43
     felicidade
    0.43
     tranquilidad
    0.42
     güz
    0.42
    ু
    0.42
    Act Density 0.025%

    No Known Activations