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

    Neuronpedia

    INDEX
    Explanations

    zero or nonzero values

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     घ
    0.47
     Finn
    0.45
     റ
    0.43
    เข
    0.42
     Terrace
    0.41
     W
    0.41
     DIRE
    0.41
     tumours
    0.41
     G
    0.41
     ischa
    0.40
    POSITIVE LOGITS
    vamente
    0.46
    kette
    0.43
    лись
    0.42
    astern
    0.41
    んば
    0.41
    ная
    0.40
    ină
    0.40
    afstand
    0.40
    nonzero
    0.39
    ná
    0.39
    Act Density 0.061%

    No Known Activations