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

    Neuronpedia

    INDEX
    Explanations

    concurrent operations and I/O

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     streaming
    -1.09
    streaming
    -1.05
     Streaming
    -1.04
    Streaming
    -0.92
     streamed
    -0.83
    照
    -0.83
     Vib
    -0.79
     Streams
    -0.77
    били
    -0.77
    zor
    -0.75
    POSITIVE LOGITS
     Faj
    0.91
    Imb
    0.80
     Loreto
    0.80
    nija
    0.76
    recette
    0.75
     fail
    0.75
     sentry
    0.74
    sentry
    0.73
     failure
    0.73
     rd
    0.71
    Act Density 0.036%

    No Known Activations