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

    Neuronpedia

    INDEX
    Explanations

    }

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     NE
    -0.07
     SW
    -0.07
    embers
    -0.07
     fram
    -0.07
     Refresh
    -0.07
     Lomb
    -0.06
    39
    -0.06
    392
    -0.06
     toàn
    -0.06
    multip
    -0.06
    POSITIVE LOGITS
    /dir
    0.08
    	trace
    0.07
    ={!
    0.07
    _Syntax
    0.06
    <k
    0.06
    .Linked
    0.06
     Luke
    0.06
    .react
    0.06
    _STAT
    0.06
    .psi
    0.06
    Act Density 0.016%

    No Known Activations