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

    Neuronpedia

    INDEX
    Explanations

    dimension / dimensions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     estrema
    -2.97
     trasparente
    -2.84
     doppia
    -2.67
     sicura
    -2.66
     When
    -2.66
    棡
    -2.63
     piú
    -2.61
     aggiunge
    -2.53
     svolge
    -2.53
     überprü
    -2.50
    POSITIVE LOGITS
    .
    3.19
      
    2.63
     –
    2.52
    而后
    2.44
     (
    2.41
    ⟞
    2.39
    <em>
    2.38
     &
    2.30
     of
    2.28
    ”،
    2.28
    Act Density 0.017%

    No Known Activations