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

    Neuronpedia

    INDEX
    Explanations

    divisions or parts

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     layer
    0.86
     слой
    0.82
     препаратов
    0.82
     సినిమాలు
    0.80
     quilts
    0.79
     impianti
    0.79
     tumours
    0.78
     loans
    0.76
    layer
    0.74
     infecciones
    0.73
    POSITIVE LOGITS
     phases
    1.72
     stages
    1.57
     sides
    1.55
    phases
    1.52
     Stages
    1.51
     sexes
    1.47
     Phases
    1.43
     eras
    1.43
    Stages
    1.42
     halves
    1.38
    Act Density 0.683%

    No Known Activations