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

    Neuronpedia

    INDEX
    Explanations

    words related to physical movements and gestures

    oai_token-act-pair · gpt-3.5-turbo

    single-letter words and short prefixes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    quartered
    -0.78
     extracts
    -0.68
     substitutes
    -0.64
     promot
    -0.64
     learners
    -0.64
     Annotations
    -0.61
     abducted
    -0.60
     modelling
    -0.60
     sacrific
    -0.60
     rescued
    -0.59
    POSITIVE LOGITS
    itty
    1.07
    agging
    0.99
    anky
    0.97
    ippy
    0.94
    inky
    0.89
    iggle
    0.88
    ooky
    0.88
    izzle
    0.87
    ummy
    0.87
    umb
    0.86
    Act Density 0.135%

    No Known Activations