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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     consistent
    -0.08
     capability
    -0.07
    ター
    -0.07
     capturing
    -0.07
    ื
    -0.07
    回
    -0.07
    ತರ
    -0.07
     solving
    -0.07
     contaminants
    -0.07
    Cons
    -0.07
    POSITIVE LOGITS
     congratulate
    0.10
     exclaimed
    0.10
     склон
    0.10
     fprintf
    0.09
    Talking
    0.09
     Pron
    0.09
     noun
    0.08
     ആഘ
    0.08
     Veter
    0.08
     Congrats
    0.08
    Act Density 0.002%

    No Known Activations