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

    Neuronpedia

    INDEX
    Explanations

    measures or quantifies

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     و
    0.73
     и
    0.70
    amazing
    0.67
    และ
    0.63
    balance
    0.61
    review
    0.58
    Balance
    0.57
    quem
    0.57
    rock
    0.57
    ᵏ
    0.57
    POSITIVE LOGITS
    \}.
    0.57
     інших
    0.56
     detained
    0.54
     ofthe
    0.54
     pula
    0.52
     spécifique
    0.51
     explicitly
    0.51
     specifically
    0.50
     deduced
    0.50
     이러한
    0.50
    Act Density 0.143%

    No Known Activations