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

    Neuronpedia

    INDEX
    Explanations

    how-questions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     what
    0.72
     best
    0.71
     Best
    0.63
     campsites
    0.62
     dinner
    0.62
     autograph
    0.61
     rendu
    0.61
    签名
    0.61
    Best
    0.61
     justice
    0.61
    POSITIVE LOGITS
     gebruiken
    0.92
    To
    0.86
     Do
    0.83
     To
    0.83
    Do
    0.82
     verwenden
    0.80
     utilizzare
    0.78
    lat
    0.77
     사용하는
    0.76
    Improving
    0.76
    Act Density 0.051%

    No Known Activations