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

    Neuronpedia

    INDEX
    Explanations

    producing various outputs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     створення
    0.55
     δημιουργ
    0.55
     creare
    0.53
     creato
    0.52
    PACKAGE
    0.52
     использования
    0.51
     scrollBody
    0.51
     creazione
    0.51
     создания
    0.50
     sfrutt
    0.50
    POSITIVE LOGITS
     output
    0.82
     outputs
    0.77
     produce
    0.71
    output
    0.70
     results
    0.69
     produces
    0.68
     Output
    0.68
     emits
    0.65
     Outputs
    0.65
     результатов
    0.64
    Act Density 0.069%

    No Known Activations