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

    Neuronpedia

    INDEX
    Explanations

    technical writing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Line
    0.43
     thousands
    0.40
     line
    0.39
     線
    0.39
    chte
    0.38
    線
    0.38
     LINE
    0.38
     hundreds
    0.37
    лло
    0.37
     number
    0.37
    POSITIVE LOGITS
     sigmoid
    0.43
    掻
    0.42
     optic
    0.41
    亚太
    0.41
    apellido
    0.40
    公告
    0.39
     động
    0.39
    ‍
    0.39
     spinel
    0.38
    Detach
    0.38
    Act Density 0.002%

    No Known Activations