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

    Neuronpedia

    INDEX
    Explanations

    logarithms, variables

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    捍卫
    -0.07
     propel
    -0.07
    Phones
    -0.07
    .C
    -0.07
     hinges
    -0.06
     Shine
    -0.06
    犇
    -0.06
     Fate
    -0.06
    (className
    -0.06
     Stokes
    -0.06
    POSITIVE LOGITS
    Array
    0.07
    target
    0.07
     Respect
    0.07
    ollar
    0.07
     Carlo
    0.07
     Rigidbody
    0.07
     clique
    0.07
    Fault
    0.07
     Divide
    0.07
    슌
    0.07
    Act Density 0.004%

    No Known Activations