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

    Neuronpedia

    INDEX
    Explanations

    Nullable

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     centroids
    -0.07
     ين
    -0.07
     jun
    -0.07
     fists
    -0.07
     wind
    -0.07
     festive
    -0.07
     Canal
    -0.07
    üs
    -0.06
     trails
    -0.06
    ’a
    -0.06
    POSITIVE LOGITS
     　　　　
    0.07
    gy
    0.07
    ammer
    0.06
     Swagger
    0.06
    ,
    0.06
    .fb
    0.06
     poisoned
    0.06
     profile
    0.06
     Alphabet
    0.06
    ]=>
    0.06
    Act Density 0.001%

    No Known Activations