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

    Neuronpedia

    INDEX
    Explanations

    recognized sooner

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    FromBody
    0.42
     refrain
    0.38
    ージャ
    0.38
    czenie
    0.35
     社会
    0.35
     overloaded
    0.34
    的精神
    0.34
    тена
    0.34
    基层
    0.33
    不在
    0.33
    POSITIVE LOGITS
     Madison
    0.46
    Madison
    0.42
    rock
    0.42
    eating
    0.40
     perfe
    0.40
     perfecting
    0.39
     país
    0.38
    ρώ
    0.37
    ап
    0.37
    bapt
    0.37
    Act Density 0.003%

    No Known Activations