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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating the completion or availability of resources or tasks

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    pine
    -0.74
    oshi
    -0.71
    tip
    -0.68
    uss
    -0.63
    Enlarge
    -0.62
    azar
    -0.62
    ike
    -0.62
    hal
    -0.61
    adin
    -0.60
    etting
    -0.59
    POSITIVE LOGITS
     plenty
    1.09
     access
    1.07
     ample
    1.05
     recourse
    1.04
     mastered
    1.00
     perfected
    0.92
     sufficient
    0.91
     solved
    0.88
     gotten
    0.87
     been
    0.87
    Act Density 0.232%

    No Known Activations