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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating processes or transitions across a range or timeline

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     alone
    -0.31
     or
    -0.31
     are
    -0.31
    onError
    -0.30
     Har
    -0.29
     Bren
    -0.29
     loss
    -0.29
     they
    -0.28
     otherwise
    -0.28
     trade
    -0.27
    POSITIVE LOGITS
    帖最后由
    0.83
     noDo
    0.79
     nahilalakip
    0.78
    <unused16>
    0.77
    <unused43>
    0.77
    <unused3>
    0.77
    <unused17>
    0.77
    fjspx
    0.77
    <unused7>
    0.76
    <unused8>
    0.76
    Act Density 0.441%

    No Known Activations