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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating something that is out of someone's control, often related to challenges, abilities, or circumstances

    oai_token-act-pair · gpt-3.5-turbo

    elements that indicate limitations or boundaries, particularly those that are perceived as beyond control or reach

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    onna
    -0.72
     differently
    -0.70
    another
    -0.70
     Marketable
    -0.69
    dylib
    -0.69
     followed
    -0.67
     anew
    -0.66
    md
    -0.66
    DAY
    -0.65
    deen
    -0.64
    POSITIVE LOGITS
     bounds
    1.40
     confines
    1.29
     boundaries
    1.20
     scope
    1.11
     comprehension
    1.07
     limits
    1.05
     reach
    1.03
     horizon
    1.00
     borders
    1.00
     realm
    0.98
    Act Density 0.137%

    No Known Activations