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

    Neuronpedia

    INDEX
    Explanations

    table summarizing differences

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Highlights
    0.45
     Highlights
    0.38
    근
    0.37
     JAMES
    0.35
     trenut
    0.33
     notas
    0.33
    highlights
    0.33
    Deferred
    0.32
    脳
    0.32
     روایت
    0.32
    POSITIVE LOGITS
     '|')
    0.40
    -|
    0.39
     "|
    0.36
     запу
    0.35
    _|
    0.35
    ῃ
    0.34
    ेटेड
    0.34
    ┆
    0.34
     Reinforced
    0.33
    etric
    0.33
    Act Density 0.003%

    No Known Activations