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

    Neuronpedia

    INDEX
    Explanations

    new features or additions mentioned in software updates

    oai_token-act-pair · gpt-3.5-turbo

    items or features related to updates and new additions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    osate
    -0.83
    oche
    -0.75
    abet
    -0.75
    ppe
    -0.74
    aminer
    -0.72
    hammad
    -0.71
    SPONSORED
    -0.71
    otle
    -0.70
    geries
    -0.68
    ontent
    -0.68
    POSITIVE LOGITS
    bies
    1.42
    bie
    1.24
    foundland
    1.02
     additions
    1.00
     arrivals
    0.98
     batch
    0.85
     entrants
    0.84
     Zealand
    0.82
     YORK
    0.81
    castle
    0.81
    Act Density 0.093%

    No Known Activations