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

    Neuronpedia

    INDEX
    Explanations

    alphanumeric characters that are part of a specific pattern, predominantly involving special characters

    oai_token-act-pair · gpt-3.5-turbo

    nested structures or brackets in mathematical or programming contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     corrid
    -0.85
     pastry
    -0.76
     Canal
    -0.70
     reeling
    -0.70
    eper
    -0.68
    chnology
    -0.66
     eaves
    -0.65
    arty
    -0.64
     needles
    -0.64
     conclud
    -0.64
    POSITIVE LOGITS
    },
    0.95
    !--
    0.92
    self
    0.85
    };
    0.85
    }{
    0.85
    });
    0.84
    aligned
    0.83
    trans
    0.82
    })
    0.82
     actionGroup
    0.80
    Act Density 0.011%

    No Known Activations