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

    Neuronpedia

    INDEX
    Explanations

    correctness

    np_max-act · gemini-2.0-flash

    the adverb "correctly" and related words about accuracy or correctness of answers and performance.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @faiazazmain99
    New Auto-Interp
    Negative Logits
    istributed
    -0.33
     directed
    -0.28
     Theater
    -0.24
    earable
    -0.24
    глав
    -0.24
    马æĭī
    -0.24
    åı°è¯į
    -0.24
    éĿ¢è²Į
    -0.24
     successful
    -0.24
    vironments
    -0.24
    POSITIVE LOGITS
    soever
    0.29
    emic
    0.27
     concent
    0.25
    ç§ģèIJ¥
    0.24
    amins
    0.24
    è¾Ľåĭ¤
    0.24
    rox
    0.24
    çijķçĸµ
    0.23
    tab
    0.23
     brakes
    0.23
    Act Density 0.003%

    No Known Activations