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

    Neuronpedia

    INDEX
    Explanations

    train lines

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    Boy
    -0.08
    icamente
    -0.07
    Document
    -0.07
    dere
    -0.07
     restored
    -0.07
    ials
    -0.07
     Like
    -0.07
     sector
    -0.06
     years
    -0.06
     Notice
    -0.06
    POSITIVE LOGITS
    แน
    0.07
    ：</
    0.06
     jsx
    0.06
    �
    0.06
    [class
    0.06
     Sweep
    0.06
    가격
    0.06
    �습니다
    0.05
     etkili
    0.05
    	esc
    0.05
    Act Density 0.019%

    No Known Activations