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

    Neuronpedia

    INDEX
    Explanations

    sentences with words that have a hyphenated prefix indicating a specific action or condition

    oai_token-act-pair · gpt-3.5-turbo

    negative prefixes and expressions related to dismemberment or removal

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    itol
    -0.74
    imum
    -0.69
    ica
    -0.66
    anni
    -0.66
    ardi
    -0.62
    ologue
    -0.62
    ère
    -0.61
    iour
    -0.61
    arf
    -0.61
    bending
    -0.61
    POSITIVE LOGITS
    ploy
    0.66
     [|
    0.64
    Downloadha
    0.62
    abase
    0.61
     manner
    0.60
    liga
    0.60
     (>
    0.60
    CLASSIFIED
    0.59
    models
    0.58
    merce
    0.58
    Act Density 0.833%

    No Known Activations