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

    Neuronpedia

    INDEX
    Explanations

    understanding and focus

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Geschä
    0.75
    ديد
    0.72
     اینکه
    0.72
     Pří
    0.72
    <unused2199>
    0.71
    बत
    0.70
     diberi
    0.70
    dracht
    0.70
    krét
    0.70
    contador
    0.68
    POSITIVE LOGITS
    .
    1.03
    .?
    0.98
    ;
    0.97
     iteratively
    0.91
     aloud
    0.89
     एवं
    0.87
    .(
    0.84
    ☀️
    0.84
    ¸
    0.84
     ولك
    0.83
    Act Density 0.701%

    No Known Activations