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

    Neuronpedia

    INDEX
    Explanations

    phrases describing challenging or distressing situations

    oai_token-act-pair · gpt-3.5-turbo

    prepositions and contextual phrases indicating relationships or conditions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ashtra
    -0.69
    nikov
    -0.66
    hov
    -0.61
     Maur
    -0.57
     IOC
    -0.57
     ASA
    -0.57
     STOP
    -0.56
    ãĤĬ
    -0.56
     resumed
    -0.55
     DEA
    -0.54
    POSITIVE LOGITS
    unch
    0.68
    ombs
    0.65
    rouse
    0.62
    ¿½
    0.61
    ruce
    0.60
    days
    0.60
    efficient
    0.59
     regard
    0.59
    arching
    0.58
    reason
    0.57
    Act Density 0.434%

    No Known Activations