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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "not" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     soient
    -0.85
    Be
    -0.79
     puissent
    -0.76
     Be
    -0.75
    be
    -0.73
    BE
    -0.66
    would
    -0.64
     possano
    -0.63
     siano
    -0.61
     būtų
    -0.60
    POSITIVE LOGITS
     need
    1.25
     have
    1.21
     want
    1.19
     deserve
    1.18
     belong
    1.18
     seem
    1.16
     owe
    1.10
     know
    1.01
     require
    0.93
     appear
    0.93
    Act Density 0.201%

    No Known Activations