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

    Neuronpedia

    INDEX
    Explanations

    quant

    np_max-act · gemini-2.0-flash

    The neuron activates on words derived from the “quant-” root (quantify, quantification, quantitative, quantitated, etc.), i.e. terms indicating quantitative measurement.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    นคร
    -0.07
    (StringUtils
    -0.07
    _probs
    -0.07
     Arcade
    -0.07
     fathers
    -0.06
     Blues
    -0.06
     Feedback
    -0.06
    _CTL
    -0.06
    _Blue
    -0.06
    اهرة
    -0.06
    POSITIVE LOGITS
    �
    0.07
     visually
    0.07
     parach
    0.07
    ]↵↵↵↵
    0.07
    Stencil
    0.06
     summ
    0.06
    )?;↵↵
    0.06
     monoc
    0.06
     Nicar
    0.06
    ζ
    0.06
    Act Density 0.004%

    No Known Activations