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

    Neuronpedia

    INDEX
    Explanations

    purpose is to be helpful

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     повто
    0.64
    再次
    0.63
     Source
    0.63
    արդ
    0.62
     Drama
    0.62
    😨
    0.61
     drama
    0.61
    ダブル
    0.60
     Performing
    0.60
     زبان
    0.59
    POSITIVE LOGITS
     ajudar
    0.85
    ebo
    0.80
     volunte
    0.77
     volunteered
    0.77
     设计
    0.76
     servent
    0.76
    惚
    0.76
     help
    0.75
     servidor
    0.75
     proteger
    0.75
    Act Density 0.026%

    No Known Activations