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

    Neuronpedia

    INDEX
    Explanations

    discussing risks and vulnerabilities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     koc
    0.45
    のデザイン
    0.42
    stripe
    0.42
    ități
    0.40
    ന്ത്രി
    0.39
     체크
    0.39
     ká
    0.39
     mape
    0.39
    wski
    0.39
    铳
    0.39
    POSITIVE LOGITS
    料
    0.41
     Delivered
    0.38
     fading
    0.38
     EN
    0.36
     solvation
    0.36
     That
    0.35
     rasp
    0.34
     deception
    0.34
    мся
    0.34
     roadside
    0.33
    Act Density 0.000%

    No Known Activations