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

    Neuronpedia

    INDEX
    Explanations

    models, drivers, or policy

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     страхо
    0.55
    ьи
    0.52
     хотите
    0.52
     և
    0.51
    하지만
    0.50
     ತೋ
    0.50
     ме
    0.49
     صہیونیوں
    0.49
     চালিয়ে
    0.49
     Санкт
    0.48
    POSITIVE LOGITS
     owls
    0.57
    ↵
    0.54
     birdie
    0.52
    Password
    0.50
     at
    0.49
    birds
    0.49
     elves
    0.48
     on
    0.47
     oiseaux
    0.47
    日
    0.46
    Act Density 0.002%

    No Known Activations