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

    Neuronpedia

    INDEX
    Explanations

    for [words/entities]

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    in
    0.79
    I
    0.52
    es
    0.51
    Tony
    0.50
    𝗼
    0.50
    ewöhn
    0.50
    oč
    0.49
     ktoré
    0.49
    aile
    0.47
    Frames
    0.47
    POSITIVE LOGITS
     रवाना
    0.48
     asuntos
    0.46
     naught
    0.45
    бав
    0.45
    geries
    0.45
     جب
    0.44
    جن
    0.44
     sake
    0.43
    すると
    0.42
    VRDR
    0.42
    Act Density 0.011%

    No Known Activations