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

    Neuronpedia

    INDEX
    Explanations

    avoid modifying

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ka
    0.88
    ali
    0.84
    7
    0.81
    na
    0.80
    ak
    0.76
    ku
    0.73
    start
    0.70
    bi
    0.70
    circledR
    0.69
    ᱗
    0.68
    POSITIVE LOGITS
    я
    0.88
    ம்
    0.78
     modifié
    0.73
    عين
    0.71
    ం
    0.70
    ות
    0.68
    ור
    0.68
     dạng
    0.66
    ون
    0.66
    ونات
    0.66
    Act Density 0.038%

    No Known Activations