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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating possibility or necessity

    oai_token-act-pair · gpt-3.5-turbo

    phrases suggesting uncertainty or speculation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ĸļ
    -0.87
    llah
    -0.65
    always
    -0.60
     relentlessly
    -0.60
     è£ıç
    -0.60
     tirelessly
    -0.60
    everything
    -0.59
    every
    -0.59
    peror
    -0.57
    ulner
    -0.56
    POSITIVE LOGITS
     someday
    1.40
     depending
    0.94
     momentarily
    0.92
     slightly
    0.91
     somew
    0.87
     some
    0.87
     unintentionally
    0.87
     slight
    0.87
     temporarily
    0.87
     inadvertently
    0.86
    Act Density 0.605%

    No Known Activations