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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating exclusivity or uniqueness

    oai_token-act-pair · gpt-3.5-turbo

    phrases emphasizing exclusivity or singularity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    storms
    -0.74
    des
    -0.71
    ence
    -0.71
    put
    -0.70
    etz
    -0.70
    mas
    -0.69
    redits
    -0.69
    ruary
    -0.68
    rs
    -0.68
    dp
    -0.67
    POSITIVE LOGITS
     thing
    1.18
     conceivable
    1.16
     reason
    1.14
     remaining
    1.11
     exception
    1.08
     way
    1.04
     drawback
    0.99
     difference
    0.98
     real
    0.98
     viable
    0.95
    Act Density 0.051%

    No Known Activations