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

    Neuronpedia

    INDEX
    Explanations

    words related to persistence or extreme intensity

    oai_token-act-pair · gpt-3.5-turbo

    adjectives that describe extreme or exaggerated characteristics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    achev
    -0.74
    ocaust
    -0.73
     Breaker
    -0.71
    oother
    -0.68
    arij
    -0.68
    atur
    -0.68
    Downloadha
    -0.67
     largeDownload
    -0.65
    bath
    -0.65
    foreseen
    -0.64
    POSITIVE LOGITS
    nesses
    1.12
    ness
    1.07
    Magikarp
    0.75
    ures
    0.74
     understatement
    0.73
     curiosity
    0.71
     coincidence
    0.71
    lihood
    0.70
    ity
    0.69
     altru
    0.69
    Act Density 0.070%

    No Known Activations