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

    Neuronpedia

    INDEX
    Explanations

    words related to actions or processes such as 'store', 'document', 'shield', 'battle', 'host', 'test', 'scale', 'judge', 'filter', 'form', 'sanction', 'spot', 'ship'

    oai_token-act-pair · gpt-3.5-turbo

    actions related to documenting, reporting, and influencing decisions or events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    pmwiki
    -0.73
    fml
    -0.68
    ritical
    -0.68
     guiActiveUnfocused
    -0.66
    sm
    -0.64
    catentry
    -0.64
    spot
    -0.62
     afore
    -0.62
    BuyableInstoreAndOnline
    -0.62
     vibe
    -0.61
    POSITIVE LOGITS
    uate
    0.98
     him
    0.84
     them
    0.82
    ulate
    0.79
     us
    0.73
    olate
    0.71
    lessly
    0.68
    inately
    0.68
    UGE
    0.68
    ISE
    0.67
    Act Density 0.853%

    No Known Activations