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

    Neuronpedia

    INDEX
    Explanations

    phrases related to selecting or choosing options from a list or group

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating selection or choice from specified groups or categories

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    thing
    -0.72
    mouth
    -0.69
    _-
    -0.65
    apo
    -0.65
    mination
    -0.64
    DEBUG
    -0.64
    quote
    -0.63
    cule
    -0.61
    rowth
    -0.61
    DIV
    -0.61
    POSITIVE LOGITS
     assorted
    0.91
     dozens
    0.86
     myriad
    0.84
     available
    0.84
     list
    0.83
     innumerable
    0.82
     hundreds
    0.81
     pool
    0.80
     various
    0.79
     among
    0.79
    Act Density 0.175%

    No Known Activations