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

    Neuronpedia

    INDEX
    Explanations

    phrases related to legal actions and responsibilities, as well as discussions about truth and falsehood

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     trench
    -0.23
     trenches
    -0.22
     precedence
    -0.22
     inequality
    -0.21
     shortages
    -0.21
     desks
    -0.21
     antit
    -0.20
     lawy
    -0.20
     dips
    -0.20
     Dresden
    -0.20
    POSITIVE LOGITS
     succeed
    0.29
    artney
    0.29
    esm
    0.29
    kees
    0.29
    ignt
    0.27
     qualify
    0.27
    olit
    0.25
    cca
    0.25
     properly
    0.25
    akings
    0.25
    Act Density 11.717%

    No Known Activations