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

    Neuronpedia

    INDEX
    Explanations

    and followed by descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Và
    0.23
    ().
    0.22
    '
    0.21
    ؛
    0.20
    ，
    0.20
    weathermap
    0.19
    }
    0.19
     ឬ
    0.19
    =>
    0.18
    ،
    0.18
    POSITIVE LOGITS
    rog
    0.21
     downright
    0.18
    поте
    0.18
     frankly
    0.18
     realist
    0.17
     unapolog
    0.17
     potere
    0.17
    ди
    0.17
     fundamento
    0.17
     divertido
    0.16
    Act Density 2.162%

    No Known Activations