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

    Neuronpedia

    INDEX
    Explanations

    Journaling, Link, Brand, Continual

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     dipping
    0.50
    ্ে
    0.49
    lán
    0.45
    tained
    0.45
    箴
    0.45
     cov
    0.45
     heavyweight
    0.44
     drill
    0.44
    drill
    0.44
     eksper
    0.43
    POSITIVE LOGITS
    宇宙
    0.47
    兆
    0.46
    を見た
    0.46
    υν
    0.45
    ρυθ
    0.44
    排除
    0.43
     arada
    0.43
    ツール
    0.42
    억
    0.42
    ジア
    0.42
    Act Density 0.001%

    No Known Activations