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

    Neuronpedia

    INDEX
    Explanations

    computer vision/feature extraction

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Properties
    -0.07
    irms
    -0.07
     extends
    -0.06
    enumerator
    -0.06
     pearl
    -0.06
    /********************************************************************************
    -0.06
    ��
    -0.06
    Richard
    -0.06
    -module
    -0.06
     speeches
    -0.06
    POSITIVE LOGITS
     Desired
    0.07
     الذي
    0.07
    leground
    0.07
    ,axis
    0.06
     desenv
    0.06
     अभ
    0.06
    भग
    0.06
    薩
    0.06
    のだ
    0.06
     určitě
    0.06
    Act Density 0.009%

    No Known Activations