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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating improbability or unlikelihood

    oai_token-act-pair · gpt-3.5-turbo

    expressions of improbability or skepticism regarding future events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ravings
    -0.88
    ocked
    -0.82
    artney
    -0.82
    zeb
    -0.80
    ulative
    -0.79
    ixed
    -0.78
    atered
    -0.77
    bara
    -0.74
    insula
    -0.73
    aeper
    -0.72
    POSITIVE LOGITS
     unlikely
    0.92
     unanim
    0.83
     unanimous
    0.82
    bably
    0.82
    theless
    0.81
    icably
    0.80
     improbable
    0.77
     fortun
    0.75
     impeachment
    0.74
     infall
    0.74
    Act Density 0.007%

    No Known Activations