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

    Neuronpedia

    INDEX
    Explanations

    verbs related to the ability to perform actions

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating the ability or capability to perform actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    çīĪ
    -0.67
     Shake
    -0.62
    cycle
    -0.61
    ocide
    -0.60
     Issues
    -0.59
     Tags
    -0.59
    Mods
    -0.59
    prise
    -0.59
     Movies
    -0.59
    odd
    -0.58
    POSITIVE LOGITS
     able
    3.26
     unable
    1.83
     Able
    1.71
     capable
    1.71
     willing
    1.54
     incapable
    1.28
     aware
    1.17
     unwilling
    1.13
     obliged
    1.11
     fortunate
    1.11
    Act Density 0.032%

    No Known Activations