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

    Neuronpedia

    INDEX
    Explanations

    phrases containing the preposition "on" with a high level of attention on the word itself

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "on" followed by numerical values or lists

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     oblig
    -0.63
    ̶
    -0.60
    ν
    -0.56
     grounding
    -0.56
     rooting
    -0.56
     Pyr
    -0.55
    ÏĤ
    -0.55
     Galile
    -0.55
     selves
    -0.53
    wig
    -0.53
    POSITIVE LOGITS
    screen
    1.20
    erous
    1.15
    etime
    1.06
     behalf
    1.06
    sets
    0.94
     occasion
    0.90
    ibaba
    0.90
     Forbes
    0.88
     Pastebin
    0.83
    site
    0.81
    Act Density 0.160%

    No Known Activations