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

    Neuronpedia

    INDEX
    Explanations

    Huffman coding

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    楚
    -0.07
    罴
    -0.07
     feared
    -0.07
    fecha
    -0.07
     attrs
    -0.07
    odash
    -0.07
     nast
    -0.07
     Pierre
    -0.07
    ᄒ
    -0.06
     booze
    -0.06
    POSITIVE LOGITS
    :A
    0.08
    ICY
    0.07
    ﰼ
    0.07
    눴
    0.07
    .syn
    0.07
    )}
↵
    0.07
    requested
    0.06
     launching
    0.06
     training
    0.06
    踞
    0.06
    Act Density 0.008%

    No Known Activations