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

    Neuronpedia

    INDEX
    Explanations

    phrases describing a specific example or scenario

    oai_token-act-pair · gpt-3.5-turbo

    references to singular subjects or entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    SpaceEngineers
    -0.84
     })
    -0.74
     }}
    -0.68
     ---------
    -0.67
     });
    -0.67
    orsi
    -0.65
    }:
    -0.65
    ulk
    -0.64
    xon
    -0.63
    src
    -0.62
    POSITIVE LOGITS
     whose
    1.07
    whose
    0.87
     which
    0.86
     where
    0.84
     devoid
    0.83
     worthy
    0.82
     THAT
    0.82
     sufficiently
    0.80
     that
    0.80
     indistinguishable
    0.78
    Act Density 1.013%

    No Known Activations