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

    Neuronpedia

    INDEX
    Explanations

    breaking targets in columns

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    Quaternion
    -0.09
     festive
    -0.08
     Quaternion
    -0.08
     quaternion
    -0.08
     қос
    -0.07
     days
    -0.07
     fêtes
    -0.07
     accent
    -0.07
     fest
    -0.07
     résult
    -0.07
    POSITIVE LOGITS
     sequential
    0.10
     chronological
    0.09
    Sequential
    0.09
     Barrier
    0.09
    tier
    0.09
     क्रम
    0.09
    -ranked
    0.09
    ASCADE
    0.08
    番号
    0.08
    [level
    0.08
    Act Density 0.009%

    No Known Activations