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

    Neuronpedia

    INDEX
    Explanations

    instances where it is implied that something is likely or probable

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "it seems that" or similar variations indicating perceived observations or beliefs

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    mouth
    -0.86
    ullivan
    -0.71
    rack
    -0.69
    isine
    -0.64
    aukee
    -0.64
    andem
    -0.62
    etary
    -0.60
    zman
    -0.60
    izont
    -0.60
    UV
    -0.59
    POSITIVE LOGITS
     whoever
    0.76
     justifies
    0.64
     there
    0.62
    ratulations
    0.62
     although
    0.62
     we
    0.59
     nobody
    0.59
     someone
    0.59
     outp
    0.58
     THERE
    0.58
    Act Density 0.181%

    No Known Activations