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

    Neuronpedia

    INDEX
    Explanations

    toggle switch actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     are
    0.88
    :
    0.86
    IV
    0.85
    س
    0.81
    s
    0.80
    ។
    0.79
     in
    0.78
    ied
    0.74
    4
    0.74
    AM
    0.73
    POSITIVE LOGITS
     Toggle
    0.78
    תו
    0.75
     Allora
    0.70
     ડે
    0.70
    체가
    0.70
    isieren
    0.69
     случаи
    0.69
    适合
    0.69
     fuerte
    0.68
    ↵
    0.67
    Act Density 0.112%

    No Known Activations