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

    Neuronpedia

    INDEX
    Explanations

    phrases related to ties or connections between separate entities

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "tie" and its variations, indicating connections to competitive outcomes or relationships

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    issance
    -0.78
    perty
    -0.74
    sburg
    -0.68
    artney
    -0.68
    wagen
    -0.68
    ocate
    -0.67
    issan
    -0.67
    agher
    -0.64
    ister
    -0.64
     Mehran
    -0.63
    POSITIVE LOGITS
    tie
    1.16
    breaker
    1.13
     knot
    1.12
     knots
    0.99
    breakers
    0.99
     tie
    0.95
    red
    0.93
     breaker
    0.91
    ged
    0.88
     tying
    0.88
    Act Density 0.040%

    No Known Activations