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

    Neuronpedia

    INDEX
    Explanations

    actions or statements related to persuasion or coercion

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to motivation and decision-making

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    enery
    -0.69
     Paste
    -0.63
     Logo
    -0.63
    availability
    -0.62
    edia
    -0.61
    alde
    -0.60
    names
    -0.59
    assador
    -0.59
    etimes
    -0.58
    overed
    -0.58
    POSITIVE LOGITS
     risky
    1.02
     harder
    0.96
     undesirable
    0.92
     harsher
    0.86
     aggressive
    0.84
     certain
    0.83
     deeper
    0.81
     immoral
    0.80
     unfavorable
    0.78
    higher
    0.77
    Act Density 0.403%

    No Known Activations