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

    Neuronpedia

    INDEX
    Explanations

    groups and their arguments

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     usually
    1.70
    Usually
    1.67
    usually
    1.66
     often
    1.64
     meestal
    1.60
     Usually
    1.59
    Often
    1.57
     semua
    1.56
    通常
    1.55
     souvent
    1.53
    POSITIVE LOGITS
     downright
    1.42
     outright
    1.34
     addirittura
    0.99
     simply
    0.99
     simplemente
    0.97
     durchaus
    0.86
     legitimately
    0.84
     genuinely
    0.81
     zelfs
    0.79
     unintentional
    0.78
    Act Density 0.234%

    No Known Activations