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

    Neuronpedia

    INDEX
    Explanations

    ambiguous word problems

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    写真
    -0.08
     dirt
    -0.07
    apanese
    -0.07
     Azul
    -0.07
     putih
    -0.07
     Illuminate
    -0.07
     starving
    -0.07
    れて
    -0.07
     сним
    -0.07
     fiber
    -0.07
    POSITIVE LOGITS
     phrase
    0.10
     rounding
    0.10
     буквально
    0.10
    Phrase
    0.10
     phr
    0.10
     literal
    0.10
    _phrase
    0.10
    (expr
    0.09
    (parsed
    0.09
     ambiguous
    0.09
    Act Density 0.113%

    No Known Activations