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

    Neuronpedia

    INDEX
    Explanations

    beginning phrases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    leder
    0.84
     занимается
    0.83
    taker
    0.82
     осно
    0.82
    %);
    0.81
    taking
    0.79
    hteet
    0.78
    taken
    0.77
     compactly
    0.77
    sibling
    0.77
    POSITIVE LOGITS
    他们
    0.71
     injectors
    0.70
     Espíritu
    0.70
     різних
    0.67
     evaluaciones
    0.66
     critics
    0.64
    残酷
    0.64
    これまで
    0.64
     fists
    0.64
    سر
    0.63
    Act Density 0.004%

    No Known Activations