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

    Neuronpedia

    INDEX
    Explanations

    negative contractions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Wayback
    0.71
     연락
    0.71
     Persever
    0.70
    ាហ
    0.69
    troops
    0.68
    േരി
    0.67
    ланд
    0.66
    Holly
    0.66
     стран
    0.66
    전히
    0.66
    POSITIVE LOGITS
    '
    1.36
    ’
    1.16
     t
    0.72
    ot
    0.68
    ''
    0.62
    ít
    0.61
    `
    0.60
     ت
    0.60
     entirely
    0.60
    ()'
    0.59
    Act Density 0.156%

    No Known Activations