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

    Neuronpedia

    INDEX
    Explanations

    asterisk

    np_max-act · gemini-2.0-flash

    This neuron detects technical terms or key phrases marked up with asterisks (e.g. *italic* or **bold**) in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     miracles
    -0.07
    (Paint
    -0.07
     it
    -0.07
     an
    -0.06
     Rd
    -0.06
     violates
    -0.06
    -way
    -0.06
     ingress
    -0.06
     possessions
    -0.06
    /stat
    -0.06
    POSITIVE LOGITS
    :**
    0.08
    _.
    0.07
    .__
    0.07
     клет
    0.07
    **:
    0.07
    .
    0.07
     kem
    0.07
    **,
    0.07
    iterate
    0.07
    ellig
    0.06
    Act Density 0.026%

    No Known Activations