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

    Neuronpedia

    INDEX
    Explanations

    non-english support forums

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     legalize
    -0.06
    j
    -0.06
     %"
    -0.06
    etal
    -0.06
    .Item
    -0.06
    �
    -0.05
    بیر
    -0.05
    '%
    -0.05
    .MapFrom
    -0.05
    ='')
    -0.05
    POSITIVE LOGITS
     이러한
    0.08
     ترب
    0.07
     wildfires
    0.07
     wildfire
    0.07
     bart
    0.07
     seh
    0.07
     여기
    0.06
     род
    0.06
     bli
    0.06
     bisexual
    0.06
    Act Density 0.005%

    No Known Activations