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

    Neuronpedia

    INDEX
    Explanations

    approximately

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     devil
    0.78
     dared
    0.68
    何度も
    0.66
    라도
    0.66
     defensive
    0.65
     dares
    0.64
     uwagę
    0.62
     unanim
    0.61
     вовсе
    0.61
     এখনই
    0.60
    POSITIVE LOGITS
     approximately
    4.48
    approximately
    4.31
    Approximately
    4.06
     Approximately
    4.05
    approx
    4.03
     aproximadamente
    3.93
     approximate
    3.89
     approx
    3.86
    Approx
    3.73
     roughly
    3.69
    Act Density 0.693%

    No Known Activations