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

    Neuronpedia

    INDEX
    Explanations

    diff and patch markers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     inter
    0.37
     അട
    0.35
    Plate
    0.35
    脍
    0.34
    Passport
    0.34
    ole
    0.33
    ляр
    0.33
    rent
    0.33
    iling
    0.33
    etic
    0.33
    POSITIVE LOGITS
    @@
    0.51
    営
    0.50
     @@
    0.50
    交代
    0.47
     (@
    0.41
    :@
    0.41
    0.41
    beek
    0.40
    竟然
    0.40
     postice
    0.40
    Act Density 0.000%

    No Known Activations