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

    Neuronpedia

    INDEX
    Explanations

    evidence

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     beled
    -0.07
    viewport
    -0.07
    aj
    -0.07
    製
    -0.07
     escalation
    -0.06
     androidx
    -0.06
    Assertions
    -0.06
    茨
    -0.06
    鉄道
    -0.06
     فرمود
    -0.06
    POSITIVE LOGITS
    !!)↵
    0.07
     |-
    0.06
    _aug
    0.06
    recover
    0.06
    -Aug
    0.06
    .extra
    0.06
    extra
    0.06
     halfway
    0.06
    め
    0.06
     kf
    0.06
    Act Density 0.004%

    No Known Activations