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

    Neuronpedia

    INDEX
    Explanations

    derivatives and equations

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     طب
    -0.07
     wil
    -0.07
    jk
    -0.07
     وزن
    -0.06
    cov
    -0.06
    eração
    -0.06
     ethn
    -0.06
     makes
    -0.06
    -my
    -0.06
    Results
    -0.06
    POSITIVE LOGITS
    ())↵↵↵
    0.07
     rotates
    0.06
     misses
    0.06
    Larry
    0.06
     provisioning
    0.06
    ]↵↵↵
    0.06
     الدم
    0.06
     oppress
    0.05
     REPLACE
    0.05
    96
    0.05
    Act Density 0.012%

    No Known Activations