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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     detach
    -0.07
    employer
    -0.07
    焊
    -0.07
    Think
    -0.06
     chooses
    -0.06
    (assert
    -0.06
    裘
    -0.06
     Nobody
    -0.06
    纛
    -0.06
    兼任
    -0.06
    POSITIVE LOGITS
    ForeignKey
    0.07
     kans
    0.07
    вел
    0.07
    ibel
    0.07
    [w
    0.06
    CELER
    0.06
     최근
    0.06
    wingConstants
    0.06
    DTV
    0.06
    rid
    0.06
    Act Density 0.021%

    No Known Activations