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

    Neuronpedia

    INDEX
    Explanations

    automatic generation and actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     tenían
    1.64
    ن
    1.60
     terão
    1.49
     extremamente
    1.48
     huesos
    1.48
     témo
    1.45
     umat
    1.43
     mundial
    1.42
     echter
    1.42
     macam
    1.42
    POSITIVE LOGITS
    да
    1.88
    ки
    1.51
    ном
    1.41
    ﺤ
    1.36
    ли
    1.35
    dır
    1.35
    ம்
    1.30
    ним
    1.29
     Те
    1.27
     Ту
    1.27
    Act Density 0.011%

    No Known Activations