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

    Neuronpedia

    INDEX
    Explanations

    professional email structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    entry
    0.38
    unary
    0.37
    ils
    0.37
    addColumn
    0.37
     $|\
    0.36
    zu
    0.36
     }_{\
    0.36
     Entry
    0.35
     lvl
    0.35
    execution
    0.35
    POSITIVE LOGITS
    Subject
    0.73
    电子邮件
    0.73
    邮件
    0.70
    Dear
    0.69
    SUBJECT
    0.66
     email
    0.65
     Subject
    0.62
     emails
    0.61
     Dear
    0.61
     ईमेल
    0.61
    Act Density 0.009%

    No Known Activations