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

    Neuronpedia

    INDEX
    Explanations

    dates in a specific format

    oai_token-act-pair · gpt-3.5-turbo

    instances of "Posted" followed by timestamps or dates

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ingly
    -0.82
    isky
    -0.78
    oller
    -0.76
    ively
    -0.76
    ivo
    -0.76
    stood
    -0.73
    ishi
    -0.71
    wit
    -0.69
    ifted
    -0.67
    vernment
    -0.65
    POSITIVE LOGITS
    escription
    0.73
    MON
    0.71
     Column
    0.67
     Comments
    0.67
     Written
    0.66
    vertis
    0.66
    itors
    0.66
    monton
    0.65
    gres
    0.64
     Secrets
    0.63
    Act Density 0.018%

    No Known Activations