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

    Neuronpedia

    INDEX
    Explanations

    write a blog post or code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    یک
    1.25
    г
    1.13
    𝒅
    1.02
    к
    1.02
    การ
    1.01
    ג
    0.99
    ai
    0.98
    ק
    0.98
    מע
    0.97
    𝒕
    0.96
    POSITIVE LOGITS
    y
    1.30
    in
    1.22
     a
    1.20
     write
    1.13
     be
    1.11
     Write
    1.10
    O
    1.04
    o
    1.03
    Write
    1.01
    W
    1.00
    Act Density 0.179%

    No Known Activations