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

    Neuronpedia

    INDEX
    Explanations

    how many or how to

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Them
    0.33
    them
    0.30
     There
    0.29
     esimerk
    0.29
     bukanlah
    0.28
     Vereinigte
    0.28
     Whatever
    0.28
     Yourself
    0.27
     Lots
    0.26
     상당
    0.26
    POSITIVE LOGITS
    soever
    0.66
     exactly
    0.55
     they
    0.53
    exactly
    0.52
    beit
    0.48
     we
    0.47
    much
    0.46
     much
    0.44
     exatamente
    0.44
     else
    0.43
    Act Density 0.255%

    No Known Activations