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

    Neuronpedia

    INDEX
    Explanations

    arguably the most

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    By
    0.55
     основные
    0.55
     By
    0.55
     by
    0.54
     основным
    0.52
     주요
    0.50
    by
    0.48
    โดย
    0.46
     توسط
    0.46
     principales
    0.43
    POSITIVE LOGITS
     misschien
    0.54
     sogar
    0.53
     probably
    0.53
     vielleicht
    0.53
     understatement
    0.51
     talvez
    0.50
     forse
    0.47
     perhaps
    0.46
     esempl
    0.45
     ispir
    0.45
    Act Density 0.006%

    No Known Activations