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

    Neuronpedia

    INDEX
    Explanations

    phrases starting with "Most" followed by numerical values

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Most" and variations in usage indicating prevalence or commonality

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     steps
    -0.63
     ent
    -0.63
     dimensions
    -0.61
     pledge
    -0.60
     repr
    -0.60
     pudding
    -0.60
     servant
    -0.59
     expression
    -0.59
    ,
    -0.58
     ver
    -0.58
    POSITIVE LOGITS
    Most
    2.99
     Most
    1.97
    Many
    1.77
    Almost
    1.75
    most
    1.74
    Usually
    1.74
    Generally
    1.68
    Often
    1.67
    Typically
    1.64
    Few
    1.63
    Act Density 0.012%

    No Known Activations