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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating upward movement or improvement

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ories
    -0.17
    adecimal
    -0.17
    anges
    -0.17
    üstü
    -0.17
    avier
    -0.16
    depend
    -0.16
    enticated
    -0.16
    ftware
    -0.16
    voj
    -0.16
    alyzed
    -0.15
    POSITIVE LOGITS
    sur
    0.25
    root
    0.24
    shot
    0.23
    rightness
    0.20
    draft
    0.20
    start
    0.20
    otre
    0.19
    standing
    0.19
    sert
    0.19
    dat
    0.19
    Act Density 0.033%

    No Known Activations