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

    Neuronpedia

    INDEX
    Explanations

    countries and names of specific places

    oai_token-act-pair · gpt-3.5-turbo

    specific geographical names and terms related to politics, history, and culture

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     unde
    -0.64
    âĢ¦âĢ¦âĢ¦âĢ¦
    -0.62
    âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦
    -0.61
     ambiguous
    -0.60
     âĻ
    -0.60
    âĶĢâĶĢâĶĢâĶĢ
    -0.59
     notation
    -0.58
    ————————
    -0.58
     conflicted
    -0.57
     "<
    -0.57
    POSITIVE LOGITS
    foundland
    0.76
    luaj
    0.71
    etheless
    0.64
    PAC
    0.63
    senal
    0.63
    ocus
    0.63
    vote
    0.62
    vertisement
    0.61
    shire
    0.61
     Orig
    0.60
    Act Density 0.431%

    No Known Activations