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

    Neuronpedia

    INDEX
    Explanations

    information related to official websites and events

    oai_token-act-pair · gpt-3.5-turbo

    titles of creative works or entertainment content

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     comparable
    -0.75
     describ
    -0.74
     toget
    -0.73
     weaker
    -0.70
     essentially
    -0.69
     virtually
    -0.67
     basically
    -0.66
     weak
    -0.66
     merit
    -0.66
    .).
    -0.65
    POSITIVE LOGITS
     Presents
    0.93
     Mayhem
    0.92
     (?,
    0.86
     Ruin
    0.85
     Goodbye
    0.84
     Designs
    0.84
     Healing
    0.84
     Ruins
    0.83
     Wonders
    0.83
     Lounge
    0.83
    Act Density 0.374%

    No Known Activations