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

    Neuronpedia

    INDEX
    Explanations

    supervis

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     works
    -0.93
     Creates
    -0.84
     creates
    -0.71
    creates
    -0.70
    Creates
    -0.64
     werkt
    -0.59
     Works
    -0.59
     supervision
    -0.58
    works
    -0.57
     create
    -0.54
    POSITIVE LOGITS
    hdashline
    0.73
    rungsseite
    0.70
    ParallelGroup
    0.66
    orghini
    0.65
    帖最后由
    0.63
    SharedDtor
    0.61
     oublié
    0.60
    styleable
    0.59
     ujednoznacz
    0.57
    orsese
    0.57
    Act Density 0.034%

    No Known Activations