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

    Neuronpedia

    INDEX
    Explanations

    phrases related to the start or beginning of events or actions

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating the concept of origin or starting points

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    irm
    -0.80
    een
    -0.69
    ura
    -0.65
    olor
    -0.65
    iar
    -0.64
    camp
    -0.63
    ril
    -0.63
    cast
    -0.63
    tti
    -0.62
    faced
    -0.62
    POSITIVE LOGITS
     afar
    1.71
     scratch
    1.53
     whence
    1.36
     inception
    1.20
     outset
    1.18
     infancy
    1.03
     beginning
    0.99
     standpoint
    0.97
     cradle
    0.96
     within
    0.96
    Act Density 0.131%

    No Known Activations