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

    Neuronpedia

    INDEX
    Explanations

    defines techniques and concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     dovrà
    0.44
    如果是
    0.43
    িকাল
    0.42
     potrà
    0.42
     கண்டிப்பாக
    0.42
     tiver
    0.41
     persiapan
    0.41
    必要があります
    0.41
    어야
    0.40
     amas
    0.40
    POSITIVE LOGITS
     whereby
    1.03
     involves
    0.94
     waarbij
    0.89
     innebär
    0.78
     wherein
    0.75
     consiste
    0.74
     позволяет
    0.71
     refers
    0.71
    Instead
    0.70
     Allows
    0.70
    Act Density 0.150%

    No Known Activations