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

    Neuronpedia

    INDEX
    Explanations

    steps, capabilities, grime, prickly

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     trovare
    0.44
     menemukan
    0.40
     withdrew
    0.39
     hitta
    0.39
     EXACT
    0.39
     जानें
    0.39
     consciously
    0.38
     exactly
    0.38
     usare
    0.38
     Gage
    0.38
    POSITIVE LOGITS
     aren
    0.68
     isn
    0.66
     seem
    0.63
     differs
    0.60
     comes
    0.55
     differ
    0.54
     bode
    0.54
    Come
    0.54
     come
    0.53
     seems
    0.53
    Act Density 0.000%

    No Known Activations