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

    Neuronpedia

    INDEX
    Explanations

    longer words with suffixes indicating a state or condition

    oai_token-act-pair · gpt-3.5-turbo

    terms related to abstract concepts and evaluation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    enegger
    -0.65
    ogie
    -0.62
    sylv
    -0.59
    ouk
    -0.56
     Noon
    -0.55
    istani
    -0.55
     Bie
    -0.54
    razen
    -0.53
     Colomb
    -0.53
     Tid
    -0.53
    POSITIVE LOGITS
     fallacy
    0.82
     (%)
    0.79
    ¶
    0.78
     :=
    0.73
     Increases
    0.73
    iveness
    0.70
     Allows
    0.70
    lessly
    0.67
     doesnt
    0.65
     =
    0.65
    Act Density 0.412%

    No Known Activations