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

    Neuronpedia

    INDEX
    Explanations

    recomp

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     slaughter
    -0.97
     Majefty
    -0.96
     Efq
    -0.96
     pleaſure
    -0.85
    abestanden
    -0.85
     protoimpl
    -0.83
     fubject
    -0.82
     Chriftian
    -0.81
    WebVitals
    -0.80
     uſe
    -0.80
    POSITIVE LOGITS
    er
    0.78
    em
    0.68
    time
    0.68
    a
    0.67
    i
    0.61
    age
    0.61
    house
    0.61
    e
    0.61
    ee
    0.60
    ed
    0.59
    Act Density 0.108%

    No Known Activations