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

    Neuronpedia

    INDEX
    Explanations

    information related to quantities, such as measurements or statistics

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating measurements or rates expressed as "per" with a numeric value

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    hess
    -0.74
    ITED
    -0.73
     Fargo
    -0.67
    çĶ°
    -0.63
    çīĪ
    -0.61
    ARDIS
    -0.61
    enegger
    -0.60
    pots
    -0.60
    ForgeModLoader
    -0.60
    ãĥĥãĥĪ
    -0.59
    POSITIVE LOGITS
     capita
    1.42
    tains
    1.20
    malink
    1.00
    ishable
    0.94
    missive
    0.94
    tained
    0.92
    tain
    0.87
     person
    0.87
    versely
    0.85
    uvian
    0.83
    Act Density 0.036%

    No Known Activations