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

    Neuronpedia

    INDEX
    Explanations

    GPT, SHA, numbers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    èrement
    0.70
    டக்க
    0.68
    حال
    0.64
    slices
    0.64
    enclose
    0.62
    opération
    0.61
     '\\;'
    0.61
    additional
    0.61
    ittal
    0.60
    fragment
    0.60
    POSITIVE LOGITS
    3
    1.36
    7
    1.21
    8
    1.20
    9
    1.18
    5
    1.18
    6
    1.17
    ۳
    1.17
    2
    1.14
    ３
    1.12
    4
    1.10
    Act Density 0.300%

    No Known Activations