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

    Neuronpedia

    INDEX
    Explanations

    normalizes and converts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     disgraceful
    0.45
     allusion
    0.43
     guérison
    0.42
     goût
    0.42
     میوز
    0.42
     güz
    0.39
     Güzel
    0.39
     pokuš
    0.38
     ветра
    0.38
    回来了
    0.38
    POSITIVE LOGITS
     convert
    0.88
     converted
    0.86
     Convert
    0.84
    Convert
    0.84
     converting
    0.79
     Converting
    0.77
    convert
    0.75
     converts
    0.75
    変換
    0.68
    转换
    0.68
    Act Density 0.450%

    No Known Activations