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

    Neuronpedia

    INDEX
    Explanations

    estimates and approximations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    各種
    0.86
     различные
    0.84
     çeşitli
    0.84
     verschiedenen
    0.82
     různých
    0.79
     verschiedene
    0.79
    各种
    0.77
    various
    0.76
     различни
    0.76
     különböző
    0.74
    POSITIVE LOGITS
     approximately
    2.28
     Approximately
    2.16
     half
    2.10
    Approximately
    2.06
     roughly
    2.06
     HALF
    1.99
     Eighty
    1.94
     eighty
    1.93
     Roughly
    1.91
     Half
    1.90
    Act Density 0.467%

    No Known Activations