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

    Neuronpedia

    INDEX
    Explanations

    code, email, URLs

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    aware
    -0.28
    èµ°äºĨ
    -0.27
     alerted
    -0.26
    åIJ¬äºĨ
    -0.26
    Skeleton
    -0.25
    comed
    -0.24
    å°ĺ
    -0.24
     pleased
    -0.24
     grounding
    -0.24
    already
    -0.24
    POSITIVE LOGITS
     transform
    0.27
    éļ¼
    0.27
    çģµ
    0.26
     transforms
    0.26
    æĹ©çĤ¹
    0.26
     EITHER
    0.24
    anko
    0.24
    åij¨åĪĬ
    0.24
    å½ĵä¸ĭ
    0.24
    ä¸ĢåĪĩ
    0.24
    Act Density 0.011%

    No Known Activations