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

    Neuronpedia

    INDEX
    Explanations

    references to things that are unusual or out of the ordinary

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "odd" and its variations, indicating a focus on abnormality or peculiarity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    jri
    -0.75
    uden
    -0.74
    Reviewed
    -0.73
    apers
    -0.72
    uberty
    -0.71
    udder
    -0.70
    zanne
    -0.69
    owder
    -0.66
    asers
    -0.66
    igate
    -0.66
    POSITIVE LOGITS
    ball
    1.29
    ities
    1.20
    balls
    1.12
     numbered
    0.94
    ity
    0.91
    yssey
    0.91
    worldly
    0.83
    numbered
    0.83
    est
    0.82
    sm
    0.79
    Act Density 0.015%

    No Known Activations