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

    Neuronpedia

    INDEX
    Explanations

    expressions indicating desire or preference for additional quantity or specific entities

    oai_token-act-pair · gpt-3.5-turbo

    expressions of desire for increased quantity or availability

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    xtap
    -0.87
    liest
    -0.74
    atan
    -0.71
     1979
    -0.67
    owe
    -0.67
    ube
    -0.67
    ĺħ
    -0.67
    oes
    -0.65
    emet
    -0.63
    Rule
    -0.63
    POSITIVE LOGITS
     than
    1.24
     efficient
    0.90
     stringent
    0.89
     Than
    0.86
     meaningful
    0.82
     aggressive
    0.82
     expensive
    0.81
     sophisticated
    0.80
    HUD
    0.80
     drastic
    0.79
    Act Density 0.146%

    No Known Activations