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

    Neuronpedia

    INDEX
    Explanations

    news outlets and platforms

    oai_token-act-pair · gpt-3.5-turbo

    references to prominent news publications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    range
    -0.79
    interstitial
    -0.74
    $.
    -0.70
    gone
    -0.64
     grounding
    -0.63
    0000000000000000
    -0.63
     pant
    -0.61
    halla
    -0.60
     Confederacy
    -0.58
    sed
    -0.58
    POSITIVE LOGITS
     understands
    1.05
     correspondent
    0.97
     reports
    0.96
     investigates
    0.96
     columnist
    0.94
     reporter
    0.94
     reported
    0.93
     publishes
    0.92
     Editorial
    0.90
     ran
    0.88
    Act Density 0.239%

    No Known Activations