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

    Neuronpedia

    INDEX
    Explanations

    words related to a decline, such as "fallen."

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of "falling" or decline in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    hett
    -0.76
    ledge
    -0.76
    usp
    -0.75
    cial
    -0.73
    cius
    -0.73
    iveness
    -0.71
    apo
    -0.70
    arin
    -0.70
    eers
    -0.69
     Gutenberg
    -0.68
    POSITIVE LOGITS
     asleep
    0.95
     fallen
    0.86
    fall
    0.85
     prey
    0.76
     dancers
    0.75
     victims
    0.73
    dylib
    0.73
     debts
    0.72
     limbs
    0.72
     victim
    0.72
    Act Density 0.005%

    No Known Activations