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

    Neuronpedia

    INDEX
    Explanations

    frequent and consistent

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ধন
    0.68
    elto
    0.67
    lında
    0.67
    ఙ
    0.66
    ckte
    0.64
    ওল
    0.64
    ർന്ന
    0.63
    apura
    0.61
    žič
    0.60
    Csvg
    0.60
    POSITIVE LOGITS
     regularly
    4.42
     frequent
    3.96
     регулярно
    3.94
     regelmäßig
    3.75
     frequently
    3.73
     repeatedly
    3.63
     Frequent
    3.59
     Regularly
    3.54
     regularmente
    3.54
     consistently
    3.52
    Act Density 1.603%

    No Known Activations