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

    Neuronpedia

    INDEX
    Explanations

    designed source

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     considérable
    0.68
    σωπ
    0.68
    にて
    0.67
     disparate
    0.65
    Apply
    0.65
     sharedInstance
    0.65
     방법에
    0.65
    蒞
    0.64
    檻
    0.63
    বাসী
    0.63
    POSITIVE LOGITS
     designed
    2.83
    designed
    2.61
     Designed
    2.58
    Designed
    2.50
     conçu
    2.13
     diseñado
    2.08
     conç
    2.00
     diseñada
    1.99
     engineered
    1.75
     aimed
    1.70
    Act Density 1.280%

    No Known Activations