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

    Neuronpedia

    INDEX
    Explanations

    states and categories

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     that
    0.68
    secret
    0.49
    ປັນ
    0.48
     iniz
    0.47
    that
    0.46
    放弃
    0.46
     autocratic
    0.46
    ס
    0.46
     Producing
    0.45
    していきます
    0.45
    POSITIVE LOGITS
    een
    0.45
    <unused2135>
    0.44
     একে
    0.43
     በጣም
    0.41
    IRC
    0.41
    ené
    0.40
     обучение
    0.40
     Обще
    0.40
    তর
    0.39
     生地
    0.39
    Act Density 0.002%

    No Known Activations