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

    Neuronpedia

    INDEX
    Explanations

    whisper

    np_max-act · gemini-2.0-flash

    The neuron never activates—it’s effectively a dead (unused) neuron.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Axes
    -0.06
     Gary
    -0.06
    chemes
    -0.06
    .gca
    -0.06
     juego
    -0.06
    bial
    -0.06
    aje
    -0.06
    Specific
    -0.06
    .paint
    -0.06
    mvc
    -0.06
    POSITIVE LOGITS
     whispered
    0.13
     whisper
    0.12
     whispers
    0.11
     Whisper
    0.10
     mentioning
    0.08
    verb
    0.07
    isper
    0.07
    ifer
    0.07
    ESP
    0.06
     góp
    0.06
    Act Density 0.005%

    No Known Activations