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

    Neuronpedia

    INDEX
    Explanations

    Version numbers and legal documents

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     of
    0.52
    ك
    0.48
     ny
    0.46
    هم
    0.46
     registers
    0.45
     unico
    0.44
    equival
    0.43
    ക്
    0.43
    都可以
    0.43
     amply
    0.43
    POSITIVE LOGITS
    の設定
    0.47
    zuoye
    0.46
    iposa
    0.44
     înviat
    0.42
    YP
    0.42
     fabs
    0.42
    ubishi
    0.41
     Fang
    0.41
    ℠
    0.39
     weighing
    0.39
    Act Density 0.001%

    No Known Activations