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

    Neuronpedia

    INDEX
    Explanations

    words related to negative experiences or disappointments

    oai_token-act-pair · gpt-3.5-turbo

    mentions of summer and processes related to upgrading or improving

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    imaru
    -0.72
     memorandum
    -0.68
     Coalition
    -0.68
     corp
    -0.64
     motion
    -0.63
     focal
    -0.63
     territ
    -0.63
     cogn
    -0.62
     jurisdiction
    -0.61
     litigation
    -0.60
    POSITIVE LOGITS
    ummer
    3.10
     Upgrade
    2.48
    Upgrade
    1.17
    lift
    1.06
    SOURCE
    0.95
     Oops
    0.94
    imil
    0.92
    RAFT
    0.91
    ungle
    0.89
    \<
    0.84
    Act Density 0.034%

    No Known Activations