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

    Neuronpedia

    INDEX
    Explanations

    `127.0.0.1` or `localhost`

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     -------------
    1.78
     =====
    1.70
    ympt
    1.70
    leq
    1.68
    -)
    1.64
    Những
    1.62
    mdash
    1.58
    $$
    1.57
    geq
    1.56
     puso
    1.54
    POSITIVE LOGITS
    PHONY
    1.90
    exe
    1.82
    ా
    1.82
    wav
    1.81
    ica
    1.73
    ico
    1.72
    !..
    1.72
    warp
    1.72
     rebuke
    1.69
    iquet
    1.61
    Act Density 0.010%

    No Known Activations