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

    Neuronpedia

    INDEX
    Explanations

    Compromises and trade-offs

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    주는
    -0.08
    .EditText
    -0.07
    ��️
    -0.06
    .fname
    -0.06
    One
    -0.06
    =W
    -0.06
     hepatitis
    -0.06
    .generate
    -0.06
     Typography
    -0.06
    انی
    -0.06
    POSITIVE LOGITS
     Escorts
    0.07
    ire
    0.07
    σω
    0.07
    �
    0.06
    usercontent
    0.06
     PIC
    0.06
     vested
    0.06
     manipulation
    0.06
     predicates
    0.06
     GMT
    0.06
    Act Density 0.052%

    No Known Activations