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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating takeovers or control

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the phrase "take over."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ero
    -0.66
    WARE
    -0.66
    Detect
    -0.63
    Forward
    -0.63
     Detect
    -0.63
    IAS
    -0.62
    area
    -0.62
    ////////////////////////////////
    -0.60
    osity
    -0.60
    Mini
    -0.60
    POSITIVE LOGITS
    drive
    0.93
    tones
    0.92
    lord
    0.85
    haul
    0.77
    irgin
    0.76
     reins
    0.75
    stocks
    0.75
    lain
    0.74
    ran
    0.72
    react
    0.71
    Act Density 0.018%

    No Known Activations