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

    Neuronpedia

    INDEX
    Explanations

    GRU and LSTM

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    icida
    0.57
    żenie
    0.56
     tweeter
    0.56
    ichtung
    0.55
    ninger
    0.54
     konserv
    0.54
    ning
    0.54
    üğ
    0.53
     offerte
    0.53
     hardcover
    0.52
    POSITIVE LOGITS
    D
    0.86
    L
    0.79
    W
    0.77
    Water
    0.69
    )=
    0.68
    S
    0.68
    Sc
    0.66
    Waters
    0.66
    y
    0.65
    H
    0.64
    Act Density 0.046%

    No Known Activations