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

    Neuronpedia

    INDEX
    Explanations

    overarching concepts and words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ``
    0.36
     सुन
    0.34
     TELL
    0.34
     交換
    0.34
     लोका
    0.32
     tends
    0.32
    意见
    0.32
    Sep
    0.31
    Nk
    0.31
    **:
    0.31
    POSITIVE LOGITS
    abundance
    0.74
    tones
    0.70
    riding
    0.68
    whelming
    0.68
    arching
    0.68
    looked
    0.63
    blown
    0.62
    reaching
    0.61
    estimation
    0.59
    lying
    0.59
    Act Density 0.019%

    No Known Activations