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

    Neuronpedia

    INDEX
    Explanations

    numbers and codes

    np_acts-logits-general · gemini-2.5-flash-lite

    markers of structure in generated text—especially section starts, sentence/paragraph boundaries, punctuation, and other formatting-like tokens.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
    introduced
    0.49
     উর্দু
    0.49
    𒊏
    0.46
    뇨
    0.45
     Фургал
    0.43
    getMainUI
    0.43
    WithFieldContext
    0.43
     između
    0.42
     ئۇ
    0.42
    ہا
    0.42
    POSITIVE LOGITS
     Sight
    0.49
     Helps
    0.48
     Encryption
    0.48
     Adverse
    0.48
     Encoding
    0.47
     Recreation
    0.47
     Ronald
    0.47
     Calm
    0.47
     Algorithm
    0.45
     Sanctuary
    0.45
    Act Density 0.029%

    No Known Activations