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

    Neuronpedia

    INDEX
    Explanations

    words indicating audience engagement and collective experiences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ÙĪÙĦا
    -0.14
    æºĸ
    -0.14
    ä»»
    -0.14
    ãĥĹãĥ¬
    -0.13
    éro
    -0.13
    oro
    -0.13
    urovision
    -0.13
    duk
    -0.13
    дÑĥ
    -0.13
     Panic
    -0.13
    POSITIVE LOGITS
     talking
    0.48
     talk
    0.46
     talked
    0.41
     Talk
    0.41
     Talking
    0.40
    talk
    0.40
    Talk
    0.40
    Talking
    0.39
    -talk
    0.37
     talks
    0.33
    Act Density 0.035%

    No Known Activations