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

    Neuronpedia

    INDEX
    Explanations

    instances where an action, condition, or quantity is significantly emphasized

    oai_token-act-pair · gpt-3.5-turbo

    the word "substantially" and other related terms indicating significant changes or impacts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    etts
    -0.68
    olan
    -0.66
     Fidel
    -0.63
     Resistance
    -0.63
    isson
    -0.63
     Victims
    -0.62
    endar
    -0.62
     Elys
    -0.62
     Butterfly
    -0.62
     Circle
    -0.61
    POSITIVE LOGITS
     ILCS
    0.93
     materially
    0.81
     substantially
    0.77
     advant
    0.76
    atility
    0.75
    orsche
    0.74
    acters
    0.73
    imilar
    0.73
    lishes
    0.72
    stantial
    0.72
    Act Density 0.004%

    No Known Activations