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

    Neuronpedia

    INDEX
    Explanations

    lowest

    np_max-act · gemini-2.0-flash

    The neuron selectively responds to floating‐point number tokens (decimal fractions) that appear (typically just before the word “common”) in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     ParameterDirection
    -0.07
     đa
    -0.07
    unistd
    -0.07
    race
    -0.07
     generates
    -0.07
     "");↵↵
    -0.06
    _stock
    -0.06
    ankan
    -0.06
    QUOTE
    -0.06
     Genius
    -0.06
    POSITIVE LOGITS
     Lower
    0.07
    .isEnabled
    0.07
     harsh
    0.06
     stiff
    0.06
     وع
    0.06
    _INST
    0.06
     �
    0.06
     Architect
    0.06
    ventus
    0.06
    charging
    0.06
    Act Density 0.001%

    No Known Activations