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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating movement or direction

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     both
    -0.23
    both
    -0.18
     all
    -0.16
    Both
    -0.15
    ledge
    -0.14
     BOTH
    -0.14
     everything
    -0.14
    thing
    -0.14
    ining
    -0.14
     intrinsic
    -0.14
    POSITIVE LOGITS
     town
    0.28
    town
    0.21
     civilization
    0.20
     darkest
    0.19
     heaven
    0.19
     camp
    0.19
     headquarters
    0.19
     bed
    0.18
     jail
    0.18
     church
    0.18
    Act Density 0.706%

    No Known Activations