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

    Neuronpedia

    INDEX
    Explanations

    was

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     candidates
    -0.87
    OGND
    -0.86
    AddTagHelper
    -0.83
     victims
    -0.82
     winners
    -0.82
    ftagPool
    -0.82
     bearers
    -0.81
     sufferers
    -0.80
     carriers
    -0.79
    rawDesc
    -0.78
    POSITIVE LOGITS
     a
    1.00
     the
    0.86
     its
    0.48
     ensayos
    0.46
     genoux
    0.44
    IUrlHelper
    0.42
     débats
    0.40
    ibly
    0.40
    라이
    0.37
    ")->
    0.37
    Act Density 0.001%

    No Known Activations