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

    Neuronpedia

    INDEX
    Explanations

    words related to reintroduction or resurfacing

    oai_token-act-pair · gpt-3.5-turbo

    instances of words related to revisiting or restoring previous conditions or states

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Harbour
    -0.73
     Manila
    -0.68
     Dominion
    -0.67
     Fields
    -0.67
     Jav
    -0.65
     Io
    -0.65
     Pap
    -0.64
     Lomb
    -0.64
    NI
    -0.64
     Pag
    -0.64
    POSITIVE LOGITS
     reintrodu
    1.14
     recapt
    1.06
     reapp
    0.94
     reinvest
    0.87
     resur
    0.86
     rehe
    0.86
     reuse
    0.83
     recon
    0.82
    resh
    0.82
    resent
    0.82
    Act Density 0.017%

    No Known Activations