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

    Neuronpedia

    INDEX
    Explanations

    ise

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    losure
    -0.06
    egers
    -0.06
    `,`
    -0.06
     statements
    -0.06
    应该
    -0.06
     evidently
    -0.05
     directive
    -0.05
     tous
    -0.05
    특별
    -0.05
    lector
    -0.05
    POSITIVE LOGITS
    ISS
    0.08
     Sanctuary
    0.08
    ARRIER
    0.07
     kok
    0.07
    험
    0.07
    ери
    0.07
    岛
    0.07
     愛
    0.07
    .ST
    0.07
    touch
    0.07
    Act Density 0.002%

    No Known Activations