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

    Neuronpedia

    INDEX
    Explanations

    effort-effectiveness evaluations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     someplace
    0.40
     ngờ
    0.37
    [,,"
    0.36
    などは
    0.36
    駄
    0.35
     solamente
    0.35
    }^{-},
    0.34
    햄
    0.34
     क्वे
    0.34
     somewhere
    0.33
    POSITIVE LOGITS
    explo
    0.44
     explo
    0.44
    ethics
    0.42
     ethics
    0.40
     Un
    0.39
     بهره
    0.38
     Explo
    0.38
     Ethics
    0.37
    Un
    0.37
     эксплуа
    0.37
    Act Density 0.044%

    No Known Activations