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

    Neuronpedia

    INDEX
    Explanations

    phrases ending in ellipses (three dots)

    oai_token-act-pair · gpt-3.5-turbo

    ellipses or trailing punctuation used for dramatic effect

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ord
    -0.82
    oresc
    -0.75
    icum
    -0.75
    odi
    -0.73
    ibility
    -0.73
    inks
    -0.73
    oit
    -0.72
    icas
    -0.72
    urat
    -0.70
    flies
    -0.69
    POSITIVE LOGITS
    âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦
    0.96
    âĢ¦âĢ¦âĢ¦âĢ¦
    0.95
    ³³³³³³³³³³³³³³³³
    0.83
    ////////////////////////////////
    0.72
    .....
    0.71
     Written
    0.67
     TRANS
    0.67
    BUT
    0.64
    wait
    0.64
     Continue
    0.64
    Act Density 0.040%

    No Known Activations