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

    Neuronpedia

    INDEX
    Explanations

    numbers with a specific format - '000' followed by varying numbers

    oai_token-act-pair · gpt-3.5-turbo

    repeated numerical patterns or counts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    heit
    -0.77
    krit
    -0.75
    icka
    -0.75
    warts
    -0.69
     weakened
    -0.65
    annel
    -0.65
    ifice
    -0.64
    archment
    -0.63
    riot
    -0.62
     strained
    -0.62
    POSITIVE LOGITS
    Hz
    0.88
    mAh
    0.88
     Hz
    0.88
     crore
    0.79
     ppm
    0.78
    +
    0.74
    Mbps
    0.74
    gp
    0.74
    aires
    0.73
    px
    0.72
    Act Density 0.083%

    No Known Activations