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

    Neuronpedia

    INDEX
    Explanations

    Language: irregular verbs

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     tranquilidad
    -0.08
    沈
    -0.08
     vejo
    -0.08
     inoc
    -0.08
    ിക്ക്
    -0.08
    ್ಯದ
    -0.08
    Torque
    -0.07
     ചികിത്സ
    -0.07
     loneliness
    -0.07
     вис
    -0.07
    POSITIVE LOGITS
     irregular
    0.13
     special
    0.11
    特殊
    0.10
    _special
    0.09
    pecial
    0.09
    special
    0.09
    .special
    0.09
     anomal
    0.09
     miscellaneous
    0.09
    .Special
    0.09
    Act Density 0.011%

    No Known Activations