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

    Neuronpedia

    INDEX
    Explanations

    phrases related to generating or having ideas

    oai_token-act-pair · gpt-3.5-turbo

    expressions of creative ideas or concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     [+
    -0.66
    ricanes
    -0.65
    istle
    -0.65
    wood
    -0.65
    administ
    -0.64
    major
    -0.64
    don
    -0.64
    cki
    -0.63
    afa
    -0.61
    DF
    -0.61
    POSITIVE LOGITS
    ually
    0.98
    ural
    0.75
    reprene
    0.69
    ual
    0.68
     salads
    0.68
     moot
    0.67
     Ideas
    0.67
    ACY
    0.66
    matic
    0.66
     ideas
    0.65
    Act Density 0.032%

    No Known Activations