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

    Neuronpedia

    INDEX
    Explanations

    number sequences and rules

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    email
    0.83
     일반
    0.76
    润
    0.76
    。(
    0.70
    ↵↵↵↵↵↵
    0.69
    >());
    0.69
     دی۔
    0.68
     типо
    0.68
    ai
    0.68
    )。
    0.68
    POSITIVE LOGITS
     revelation
    0.89
     penta
    0.85
    sentences
    0.85
    yards
    0.83
     recog
    0.80
     histamine
    0.79
     hitting
    0.78
     keen
    0.77
     dismiss
    0.77
     psoriasis
    0.76
    Act Density 0.025%

    No Known Activations