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

    Neuronpedia

    INDEX
    Explanations

    numeric thresholds or minimum requirements

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating minimum quantity or requirement thresholds

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    quit
    -0.67
     happiest
    -0.65
    cro
    -0.64
    mination
    -0.61
    ãĤ¿
    -0.61
    window
    -0.60
    rend
    -0.59
    ãĤº
    -0.59
    tions
    -0.58
    icides
    -0.58
    POSITIVE LOGITS
     partially
    1.08
     partly
    1.00
     SOME
    0.91
     temporarily
    0.83
     marginally
    0.80
     moderately
    0.80
     mildly
    0.78
     partial
    0.78
     tacit
    0.76
     superf
    0.76
    Act Density 0.041%

    No Known Activations