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

    Neuronpedia

    INDEX
    Explanations

    the word "amazing" with relatively high activation values

    oai_token-act-pair · gpt-3.5-turbo

    expressions of admiration or enthusiasm

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ©¶æ
    -0.83
    epend
    -0.82
    pai
    -0.82
    embed
    -0.77
    vere
    -0.77
    eter
    -0.76
    aper
    -0.76
    avis
    -0.74
    few
    -0.72
    enser
    -0.71
    POSITIVE LOGITS
    NESS
    0.87
     rendition
    0.86
     coincidence
    0.85
     feats
    0.85
     feat
    0.83
     amazing
    0.75
     talent
    0.75
     accomplishment
    0.75
     amounts
    0.74
     incredible
    0.71
    Act Density 0.029%

    No Known Activations