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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "connect" in various forms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    </b>
    -0.92
    <b>
    -0.87
    sia
    -0.76
    er
    -0.75
    han
    -0.73
     «
    -0.72
    </i>
    -0.68
    </h1>
    -0.67
    ram
    -0.66
     P
    -0.66
    POSITIVE LOGITS
     myſelf
    1.34
     متعلقه
    1.27
     itſelf
    1.20
     étoient
    1.18
     peindre
    1.15
    ."</
    1.10
     themſelves
    1.09
     feroit
    1.06
     perſon
    1.05
     purpoſe
    1.05
    Act Density 0.076%

    No Known Activations