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

    Neuronpedia

    INDEX
    Explanations

    spacing out letters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     rappresent
    0.44
    ສໍາ
    0.42
     dep
    0.40
    dep
    0.40
     meglio
    0.40
     hỗ
    0.40
    াক
    0.39
    許多
    0.39
     apoio
    0.39
    को
    0.39
    POSITIVE LOGITS
    ｒ
    0.48
     A
    0.42
     B
    0.42
     a
    0.42
     seemingly
    0.41
     চৈতন্য
    0.41
     P
    0.40
     vitth
    0.40
     -,
    0.39
    しくは
    0.39
    Act Density 0.027%

    No Known Activations