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

    Neuronpedia

    INDEX
    Explanations

    times when something changes or transitions to a new state

    oai_token-act-pair · gpt-3.5-turbo

    repeated use of the verb "become" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     recent
    -0.70
    onz
    -0.64
     lucky
    -0.54
     Tuesday
    -0.52
    ones
    -0.51
    metry
    -0.51
    olk
    -0.51
     Wednesday
    -0.51
     FY
    -0.51
     Monday
    -0.50
    POSITIVE LOGITS
     becomes
    3.15
     ceases
    2.10
     became
    1.93
     disappears
    1.92
     become
    1.79
     emerges
    1.75
     loses
    1.72
     evolves
    1.71
     succeeds
    1.71
     grows
    1.63
    Act Density 0.012%

    No Known Activations