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

    Neuronpedia

    INDEX
    Explanations

    avoiding certain letters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    alev
    0.49
    ankton
    0.48
    dependence
    0.47
    alin
    0.47
    ondata
    0.47
    on
    0.46
    interval
    0.45
    aling
    0.45
    intervention
    0.45
    acl
    0.44
    POSITIVE LOGITS
     Vamos
    0.54
     Camero
    0.53
     Azz
    0.51
     Mena
    0.50
     Aman
    0.50
     Iman
    0.50
     Mén
    0.49
     Física
    0.49
     pestaña
    0.49
     Laj
    0.49
    Act Density 0.002%

    No Known Activations