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

    Neuronpedia

    INDEX
    Explanations

    gradual change

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     houden
    0.71
     стар
    0.68
    स्पति
    0.66
    പാട
    0.66
     Hobson
    0.66
     Samp
    0.66
     Hona
    0.64
     Ain
    0.63
    可通过
    0.63
     Starter
    0.63
    POSITIVE LOGITS
     chipping
    1.05
     slowly
    0.94
     dismantling
    0.89
     gradually
    0.89
     tightening
    0.89
     shrinking
    0.88
     zwięks
    0.88
     eroding
    0.86
     erosion
    0.85
     확대
    0.84
    Act Density 0.420%

    No Known Activations