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

    Neuronpedia

    INDEX
    Explanations

    phrases asking questions with a focus on questioning the truth or validity of statements

    oai_token-act-pair · gpt-3.5-turbo

    be verbs indicating states of being or existence

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Nightmares
    -0.73
     Messenger
    -0.73
    Reviewer
    -0.66
    stood
    -0.64
    iggins
    -0.61
     Dreams
    -0.61
     Train
    -0.59
    izer
    -0.58
     Working
    -0.57
     Dragonbound
    -0.57
    POSITIVE LOGITS
    omorphic
    0.86
    nt
    0.85
    gur
    0.83
    ĸļ
    0.81
    hap
    0.81
    peria
    0.76
    ps
    0.73
    eret
    0.71
    fred
    0.70
     ya
    0.69
    Act Density 0.204%

    No Known Activations