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

    Neuronpedia

    INDEX
    Explanations

    phrases related to things being beyond control or capability

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating limitations or things that are beyond control or capabilities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     differently
    -0.68
    itia
    -0.67
     AGA
    -0.66
    another
    -0.64
    md
    -0.64
     anew
    -0.63
     Dele
    -0.62
     somet
    -0.62
    llor
    -0.62
    hea
    -0.61
    POSITIVE LOGITS
     bounds
    1.53
     confines
    1.40
     limits
    1.25
     scope
    1.22
     boundaries
    1.20
     comprehension
    1.11
     reach
    1.11
     realm
    1.01
     borders
    0.99
     realms
    0.99
    Act Density 0.137%

    No Known Activations