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

    Neuronpedia

    INDEX
    Explanations

    periods

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    reducible
    1.09
    reactive
    1.04
    decomposition
    1.00
    turbulence
    0.99
    recipe
    0.97
    strategic
    0.96
    deformation
    0.96
    reduction
    0.94
    quantidade
    0.94
    padding
    0.94
    POSITIVE LOGITS
     Ab
    0.87
     Air
    0.83
     Ai
    0.82
     pure
    0.81
     written
    0.80
     AB
    0.79
     hard
    0.79
     Ed
    0.77
    ’
    0.77
     Esp
    0.76
    Act Density 0.000%

    No Known Activations