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

    Neuronpedia

    INDEX
    Explanations

    product reviews/guides

    np_max-act · gemini-2.0-flash

    This neuron fires on words that introduce recommendations or instructions (e.g., “look for,” “choose,” “consider”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _NETWORK
    -0.08
    put
    -0.07
    rowser
    -0.06
    .Payload
    -0.06
    	Z
    -0.06
    .untracked
    -0.06
    _UNKNOWN
    -0.06
    CanBe
    -0.06
    Band
    -0.06
    ):
↵
    -0.06
    POSITIVE LOGITS
     Chinese
    0.07
     wooded
    0.07
    (resp
    0.07
     speculated
    0.06
    ambique
    0.06
    ilik
    0.06
     Ya
    0.06
     kvinde
    0.06
    овая
    0.06
     ortak
    0.06
    Act Density 0.042%

    No Known Activations