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

    Neuronpedia

    INDEX
    Explanations

    numerical data and structured information

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     inol
    0.42
     mitad
    0.42
    謊
    0.41
     পিটার
    0.40
     uncut
    0.40
     PI
    0.39
     நடித்த
    0.39
     குடிய
    0.38
    酘
    0.38
     half
    0.38
    POSITIVE LOGITS
    Lever
    0.39
    ৯
    0.39
     },{
    0.37
    Difference
    0.37
    Leng
    0.37
    +(
    0.36
    Revolution
    0.36
     diffère
    0.36
    面倒
    0.36
    怜
    0.35
    Act Density 0.000%

    No Known Activations