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

    Neuronpedia

    INDEX
    Explanations

    wrongful

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     dump
    -0.07
    arrison
    -0.07
    ेन
    -0.06
     bottleneck
    -0.06
     evaluate
    -0.06
    -0.06
    iseum
    -0.06
     leaks
    -0.06
    áhnout
    -0.06
     fracking
    -0.05
    POSITIVE LOGITS
     wrongful
    0.14
     currentState
    0.09
    ्व
    0.08
     hol
    0.07
    (${
    0.07
    .removeItem
    0.07
    .partial
    0.07
    datap
    0.07
    (Long
    0.07
    ْد
    0.07
    Act Density 0.001%

    No Known Activations