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

    Neuronpedia

    INDEX
    Explanations

    particular phrases related to social and political issues, likely with a critical perspective

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     incumb
    -0.63
    76561
    -0.60
     ens
    -0.59
     %%
    -0.59
     lasted
    -0.58
    CLASSIFIED
    -0.57
    0000000000000000
    -0.54
     destroys
    -0.54
    bryce
    -0.53
    llor
    -0.53
    POSITIVE LOGITS
     lieu
    1.15
    clus
    1.06
     conjunction
    1.02
    ordinate
    0.99
     regards
    0.99
    clusions
    0.96
     order
    0.95
     accordance
    0.95
     terms
    0.95
     relation
    0.93
    Act Density 10.466%

    No Known Activations