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

    Neuronpedia

    INDEX
    Explanations

    express a need or requirement

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    рт
    1.13
    ну
    1.02
    <0x80>
    1.02
    тен
    1.01
    ви
    0.98
    ૦
    0.97
     شرطونه
    0.92
    ън
    0.90
    0.90
     overflowing
    0.90
    POSITIVE LOGITS
     to
    1.80
    ل
    1.61
    ל
    1.53
    א
    1.51
    د
    1.41
    RO
    1.41
    ע
    1.41
    ↵↵
    1.40
    ES
    1.38
    IN
    1.34
    Act Density 0.194%

    No Known Activations