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

    Neuronpedia

    INDEX
    Explanations

    anatomy, biology, neural networks

    np_acts-logits-general · gemini-2.5-flash-lite

    academic or technical terms in scientific and educational contexts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    New Auto-Interp
    Negative Logits
     Somehow
    -0.84
     بسیار
    -0.83
     Though
    -0.83
     impeccable
    -0.82
    鄺
    -0.81
     oftentimes
    -0.80
     Compared
    -0.79
     While
    -0.79
    While
    -0.79
    ēm
    -0.79
    POSITIVE LOGITS
    菒
    1.03
     strikingly
    0.98
    ctically
    0.90
    five
    0.89
    eight
    0.86
    during
    0.85
    four
    0.85
    mmation
    0.84
     continually
    0.84
    finally
    0.83
    Act Density 0.011%

    No Known Activations