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

    Neuronpedia

    INDEX
    Explanations

    ="

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     accusing
    -0.08
     течение
    -0.07
     terrace
    -0.07
    דאג
    -0.07
    ᓫ
    -0.07
     Peng
    -0.07
    usting
    -0.07
     등의
    -0.07
     consequence
    -0.07
     Azerbai
    -0.07
    POSITIVE LOGITS
    反转
    0.07
     thuisontvangst
    0.07
     Blob
    0.07
    书店
    0.07
     marginRight
    0.06
     typingsJapgolly
    0.06
    カラ
    0.06
    ailles
    0.06
     souvenir
    0.06
    creat
    0.06
    Act Density 0.005%

    No Known Activations