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

    Neuronpedia

    INDEX
    Explanations

    articles

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     postfix
    -0.08
    �
    -0.07
     terug
    -0.07
     Married
    -0.07
    (paths
    -0.07
    assigned
    -0.07
     recruiters
    -0.07
     belongs
    -0.07
    iliated
    -0.07
    uestos
    -0.07
    POSITIVE LOGITS
     leak
    0.08
    臌
    0.07
    @@
    0.07
    oine
    0.06
     horizontal
    0.06
    教研
    0.06
     }}"
    0.06
    簸
    0.06
    룅
    0.06
    бой
    0.06
    Act Density 0.021%

    No Known Activations