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

    Neuronpedia

    INDEX
    Explanations

    instructions or prompts for making selections or choices

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ViewFeatures
    -0.70
    etcode
    -0.66
    ecake
    -0.64
    Goodnight
    -0.61
     Harvey
    -0.58
    documented
    -0.57
    ocumented
    -0.56
     '../
    -0.56
    Unread
    -0.56
    sApp
    -0.56
    POSITIVE LOGITS
    choose
    1.67
     choose
    1.61
    Choose
    1.59
     Choose
    1.59
     choosing
    1.59
     chooses
    1.57
     chosen
    1.53
     CHOOSE
    1.50
    choosing
    1.50
     Choosing
    1.49
    Act Density 0.158%

    No Known Activations