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

    Neuronpedia

    INDEX
    Explanations

    why explain it

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     for
    0.59
    在
    0.59
    ى
    0.59
    ين
    0.55
    Name
    0.55
     in
    0.54
    га
    0.54
    (
    0.54
    도
    0.51
    US
    0.50
    POSITIVE LOGITS
    0.56
    ０
    0.50
    0.46
    kannya
    0.44
    ↵↵
    0.40
     більш
    0.38
     Į
    0.38
     різних
    0.38
    orpion
    0.36
     حداقل
    0.36
    Act Density 0.621%

    No Known Activations