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

    Neuronpedia

    INDEX
    Explanations

    say "accident"

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Derby
    -0.07
    american
    -0.07
    bones
    -0.07
    assemble
    -0.07
    Led
    -0.07
    uttle
    -0.06
     Grass
    -0.06
    fa
    -0.06
    .format
    -0.06
    aurants
    -0.06
    POSITIVE LOGITS
     "{
    0.07
    >[]
    0.06
     println
    0.06
    levation
    0.06
    itur
    0.06
    句
    0.06
     اد
    0.06
     ボ
    0.06
    .rdf
    0.06
    (++
    0.06
    Act Density 0.164%

    No Known Activations