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

    Neuronpedia

    INDEX
    Explanations

    words or phrases preceded by an uncommon character sequence

    oai_token-act-pair · gpt-3.5-turbo

    instances of high-risk or potentially dangerous situations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    heit
    -0.89
    boro
    -0.81
    ospons
    -0.78
    abouts
    -0.76
    zona
    -0.75
    reetings
    -0.75
    kees
    -0.72
    ounce
    -0.70
    osaurus
    -0.68
     Versus
    -0.68
    POSITIVE LOGITS
    Scroll
    0.91
    However
    0.90
    Shape
    0.84
    Labour
    0.84
    SPONSORED
    0.77
     councillor
    0.76
    Scotland
    0.75
     Labour
    0.74
     mould
    0.74
    Britain
    0.73
    Act Density 0.368%

    No Known Activations