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

    Neuronpedia

    INDEX
    Explanations

    ICE

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    erval
    -0.29
    odium
    -0.28
    lemen
    -0.28
    pery
    -0.27
    assage
    -0.26
    èŀ¨
    -0.25
    ubb
    -0.24
    pes
    -0.24
     slippery
    -0.24
    mitted
    -0.24
    POSITIVE LOGITS
    åĿ³
    0.28
    itar
    0.28
    å¬
    0.28
    éģIJ
    0.27
    ango
    0.27
    å°¸
    0.27
    (handles
    0.26
    bak
    0.26
     uniform
    0.24
     widest
    0.24
    Act Density 0.008%

    No Known Activations