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

    Neuronpedia

    INDEX
    Explanations

    reducing

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     hitch
    -0.07
     Harrison
    -0.07
     Intent
    -0.07
     キャ
    -0.06
     mail
    -0.06
     stepping
    -0.06
     Maple
    -0.06
    Jason
    -0.06
     chapel
    -0.06
    NICALL
    -0.06
    POSITIVE LOGITS
     reduced
    0.14
     reduce
    0.13
    reduce
    0.12
     reducing
    0.11
     reduction
    0.11
     Reduction
    0.10
     reduces
    0.10
     Reduced
    0.10
    减
    0.10
    Reducer
    0.10
    Act Density 0.057%

    No Known Activations