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

    Neuronpedia

    INDEX
    Explanations

    tapered

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    .webkit
    -0.07
     UNS
    -0.06
     bus
    -0.06
     +%
    -0.06
     WORD
    -0.06
     Scot
    -0.06
    чески
    -0.06
     Dict
    -0.06
    因
    -0.06
    	command
    -0.06
    POSITIVE LOGITS
     tapered
    0.12
     Faul
    0.08
     taper
    0.08
     reversal
    0.07
     těž
    0.07
    İZ
    0.07
    yper
    0.07
    APS
    0.07
     فو
    0.07
    _paper
    0.07
    Act Density 0.001%

    No Known Activations