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

    Neuronpedia

    INDEX
    Explanations

    terms indicating authority or control over something

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "hold" and its variations in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    endix
    -0.76
    icycle
    -0.67
    clerosis
    -0.65
     deceive
    -0.64
     Daylight
    -0.64
     omission
    -0.63
    ologist
    -0.63
    inx
    -0.62
    osal
    -0.62
    ibel
    -0.61
    POSITIVE LOGITS
    hold
    1.10
    holders
    1.07
    holder
    1.01
    holding
    0.96
     sway
    0.94
     hold
    0.94
    held
    0.85
    erness
    0.85
     accountable
    0.80
     captive
    0.77
    Act Density 0.034%

    No Known Activations