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

    Neuronpedia

    INDEX
    Explanations

    General English text

    np_max-act · gemini-2.0-flash

    The neuron responds to tokens inside parentheses or brackets—e.g. parenthetical asides, acronyms, numbers, or quoted bits.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    FETCH
    -0.07
    ensure
    -0.07
     леч
    -0.06
    Fear
    -0.06
     Evaluate
    -0.06
     Hide
    -0.06
    олуч
    -0.06
     생활
    -0.06
    ]^
    -0.06
    _tC
    -0.06
    POSITIVE LOGITS
    .but
    0.07
     WRONG
    0.07
    (&:
    0.07
    def
    0.07
     herbs
    0.07
    _ENCODING
    0.06
    -gr
    0.06
    باب
    0.06
     swath
    0.06
     déf
    0.06
    Act Density 0.014%

    No Known Activations