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

    Neuronpedia

    INDEX
    Explanations

    try to answer

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    caster
    0.94
     separate
    0.93
     succumb
    0.91
    డం
    0.90
    事前に
    0.88
     beschrieben
    0.87
    續
    0.86
     starve
    0.83
     starving
    0.83
    tego
    0.83
    POSITIVE LOGITS
    yaw
    1.13
    anche
    1.03
     diffic
    1.02
     glu
    1.01
     gpd
    0.99
     UBND
    0.99
     terras
    0.99
     হেম
    0.98
     virtu
    0.98
     voulez
    0.98
    Act Density 0.004%

    No Known Activations