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

    Neuronpedia

    INDEX
    Explanations

    closing parentheses and brackets

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    </blockquote>
    1.07
     ice
    0.94
    ICE
    0.87
    ice
    0.86
    <blockquote>
    0.84
    ']],
    0.82
     Ice
    0.82
    Ice
    0.81
     snow
    0.79
    Snow
    0.76
    POSITIVE LOGITS
    })
    3.39
    });
    3.15
     })
    3.11
    }).
    2.97
    ])
    2.95
    })=
    2.92
    "])
    2.90
    })$
    2.89
    '])
    2.88
    \})
    2.82
    Act Density 0.238%

    No Known Activations