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

    Neuronpedia

    INDEX
    Explanations

    requires careful

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Putting
    0.18
     ….
    0.18
    ህ
    0.17
    сл
    0.17
    य
    0.17
    આ
    0.17
     putting
    0.17
    …………..
    0.17
     leto
    0.17
    \"
    0.17
    POSITIVE LOGITS
     considerable
    0.32
     layers
    0.30
     весьма
    0.30
    相当
    0.29
     enough
    0.27
     nuance
    0.27
     상당
    0.27
     상당히
    0.26
     elements
    0.26
     cukup
    0.25
    Act Density 0.962%

    No Known Activations