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

    Neuronpedia

    INDEX
    Explanations

    phrases related to actions of splitting or separating

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to splitting or separating entities or concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     HIGH
    -0.67
    BILITY
    -0.62
    geries
    -0.61
     HI
    -0.58
    odor
    -0.57
    DAY
    -0.57
    shapeshifter
    -0.57
    âĨij
    -0.56
    high
    -0.56
    reviewed
    -0.55
    POSITIVE LOGITS
     into
    0.90
     barriers
    0.87
     alliances
    0.78
     fences
    0.76
     friendships
    0.75
     marriages
    0.73
     Into
    0.69
     territ
    0.69
     partitions
    0.69
     chunks
    0.68
    Act Density 0.116%

    No Known Activations