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

    Neuronpedia

    INDEX
    Explanations

    phrases related to the capability or capacity of something or someone

    oai_token-act-pair · gpt-3.5-turbo

    phrases emphasizing the ability or potential of subjects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    cot
    -0.79
    cott
    -0.72
    bard
    -0.70
    mol
    -0.68
    alter
    -0.68
    cision
    -0.66
    meric
    -0.66
     Swe
    -0.65
    gar
    -0.64
    EE
    -0.63
    POSITIVE LOGITS
     capable
    0.93
     incapable
    0.86
    bodied
    0.84
    nesses
    0.82
     accompl
    0.79
    sonian
    0.78
    ointed
    0.77
    icient
    0.76
     enough
    0.74
    ansas
    0.70
    Act Density 0.008%

    No Known Activations