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

    Neuronpedia

    INDEX
    Explanations

    Microsoft.Extensions namespaces

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ands
    0.58
    </h2>
    0.57
    ſ
    0.57
    </code>
    0.56
     tricky
    0.56
     typu
    0.56
     joka
    0.55
     pounds
    0.55
    (
    0.55
     आहे
    0.54
    POSITIVE LOGITS
    ר
    0.85
    র
    0.80
    רת
    0.80
    ر
    0.79
    น
    0.78
    ו
    0.76
    ل
    0.75
    ن
    0.75
    و
    0.73
    మ
    0.71
    Act Density 0.000%

    No Known Activations