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

    Neuronpedia

    INDEX
    Explanations

    :

    np_max-act · gemini-2.0-flash

    The neuron is effectively “dead”—it never activates (i.e. it does not pick out any tokens).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    20
    -0.07
    212
    -0.07
     taboo
    -0.06
     Serve
    -0.06
    intro
    -0.06
     IQ
    -0.06
    19
    -0.06
     Haiti
    -0.06
    %i
    -0.06
     Influ
    -0.06
    POSITIVE LOGITS
    _references
    0.08
    [B
    0.07
     specifier
    0.07
    เกษตร
    0.07
    LAN
    0.07
    เศ
    0.06
    UGINS
    0.06
     escorted
    0.06
    -single
    0.06
    นน
    0.06
    Act Density 0.006%

    No Known Activations