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

    Neuronpedia

    INDEX
    Explanations

    parts of words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ­
    0.71
     socalled
    0.71
     fi
    0.69
    ­
    0.60
     speci
    0.58
     wellknown
    0.57
     con
    0.54
    fi
    0.53
    ־
    0.53
     •
    0.50
    POSITIVE LOGITS
    https
    0.46
    "]}
    0.39
     https
    0.38
    keyValue
    0.38
    <i>
    0.38
     unre
    0.36
     })
    0.36
    ﺙ
    0.36
     }:
    0.36
     mislead
    0.35
    Act Density 0.001%

    No Known Activations