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

    Neuronpedia

    INDEX
    Explanations

    comparisons of quantities or numbers

    oai_token-act-pair · gpt-3.5-turbo

    appearances of the word "few" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Maze
    -0.68
    wrapper
    -0.68
     Bust
    -0.67
    ansion
    -0.61
    ilon
    -0.61
     Remastered
    -0.60
     Hipp
    -0.59
    kok
    -0.58
    ACTION
    -0.56
     Higher
    -0.55
    POSITIVE LOGITS
    est
    1.22
    er
    0.91
    eenth
    0.91
    ever
    0.88
    eties
    0.82
    eric
    0.82
     exceptions
    0.78
    een
    0.78
     mortals
    0.77
    eners
    0.76
    Act Density 0.043%

    No Known Activations