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

    Neuronpedia

    INDEX
    Explanations

    phrases related to duration or continuity

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to the concept of staying or remaining in a particular state or location

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Ļ½
    -0.81
     Guilty
    -0.77
    ogyn
    -0.75
    haps
    -0.74
    uesday
    -0.73
    anca
    -0.72
    intent
    -0.72
    apest
    -0.71
     Archdemon
    -0.68
    repre
    -0.68
    POSITIVE LOGITS
     indefinitely
    1.14
    lishes
    0.93
     intact
    0.86
     forever
    0.86
     confines
    0.83
    until
    0.81
     uninterrupted
    0.80
    til
    0.78
     fold
    0.78
     until
    0.78
    Act Density 0.344%

    No Known Activations