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

    Neuronpedia

    INDEX
    Explanations

    phrases related to performing actions one at a time

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate frequency and timing

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    drawn
    -0.85
    leased
    -0.76
    utters
    -0.74
    roth
    -0.74
    erest
    -0.73
    çĶ°
    -0.71
    redit
    -0.70
    ships
    -0.67
    wr
    -0.66
    vik
    -0.65
    POSITIVE LOGITS
     istg
    0.72
    secut
    0.71
     perimeter
    0.68
     rotation
    0.68
     ratio
    0.68
     gallon
    0.65
     rectangle
    0.63
     average
    0.62
     metric
    0.62
     sandwich
    0.62
    Act Density 0.156%

    No Known Activations