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

    Neuronpedia

    INDEX
    Explanations

    project paths with slashes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    _;
    0.98
    AtPath
    0.78
    _",
    0.78
     $_{
    0.75
    >`,
    0.74
    }{$
    0.73
    _
    0.72
    ?;
    0.72
    _{*}\
    0.71
    ocom
    0.71
    POSITIVE LOGITS
     Fly
    0.82
     flyer
    0.79
     Flyer
    0.79
     pisos
    0.77
     Norm
    0.75
     Scalar
    0.73
     યાદ
    0.73
     Rolle
    0.73
     Spectre
    0.72
     Seen
    0.72
    Act Density 0.029%

    No Known Activations