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

    Neuronpedia

    INDEX
    Explanations

    space, query, healthy

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    𝐬
    0.54
     აღმასრულ
    0.52
     እንዲሁ
    0.49
     フランス
    0.48
    entra
    0.48
     नशे
    0.47
     សូម
    0.47
    𝐦
    0.47
    <unused646>
    0.47
     drugi
    0.46
    POSITIVE LOGITS
     and
    0.57
     as
    0.46
     appeal
    0.46
     و
    0.44
            
    0.44
     barriers
    0.43
    Appeal
    0.42
    0.42
     appearance
    0.42
     duplicates
    0.42
    Act Density 0.001%

    No Known Activations