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

    Neuronpedia

    INDEX
    Explanations

    numbers ending in 6 or 2

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ت
    1.85
    تهم
    1.62
     demean
    1.48
    т
    1.47
     averaging
    1.46
    й
    1.43
    น์
    1.40
    是很
    1.38
    是从
    1.33
     downside
    1.32
    POSITIVE LOGITS
    een
    2.28
    ní
    1.82
    卦
    1.76
    ுக
    1.73
    ्स
    1.71
    cle
    1.70
    во
    1.69
    ো
    1.67
    ieth
    1.55
    eenth
    1.55
    Act Density 0.313%

    No Known Activations