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

    Neuronpedia

    INDEX
    Explanations

    Originally designed, conceived, comprised

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     potrebbero
    0.71
    なら
    0.70
     puedan
    0.70
     puissent
    0.67
     Losses
    0.66
     losses
    0.66
     puisse
    0.65
     mogli
    0.62
     وكذلك
    0.61
     olacaktır
    0.60
    POSITIVE LOGITS
     conçu
    1.15
     brainchild
    1.15
     headquartered
    1.13
     designed
    1.10
     ursprünglich
    1.07
     Originally
    1.04
     originally
    1.03
    它的
    1.01
    旨在
    1.01
     comprised
    0.99
    Act Density 1.805%

    No Known Activations