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

    Neuronpedia

    INDEX
    Explanations

    phrases related to compatibility or fitting concepts and ideas together

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    xba
    -0.16
    以æĿ¥
    -0.15
    istan
    -0.14
     affairs
    -0.14
     tent
    -0.14
    cba
    -0.14
    دارÛĮ
    -0.14
     fur
    -0.14
    pio
    -0.13
    vik
    -0.13
    POSITIVE LOGITS
     perfectly
    0.33
     nicely
    0.27
     PERF
    0.22
     well
    0.20
     harmon
    0.19
    perfect
    0.18
     Harmon
    0.18
     snug
    0.18
     nice
    0.18
     Perfect
    0.18
    Act Density 0.090%

    No Known Activations