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

    Neuronpedia

    INDEX
    Explanations

    diversification, reducing risk

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    θή
    -0.07
     přij
    -0.06
     moet
    -0.06
     Spor
    -0.06
    .secret
    -0.06
    _cl
    -0.06
     steals
    -0.06
     přid
    -0.06
    _WORD
    -0.06
     sondern
    -0.06
    POSITIVE LOGITS
    ography
    0.07
    ライト
    0.07
    روس
    0.06
    ')↵↵
    0.06
     freed
    0.06
     specifically
    0.06
     egal
    0.06
    ้วย
    0.06
     english
    0.06
    	ar
    0.06
    Act Density 0.009%

    No Known Activations