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

    Neuronpedia

    INDEX
    Explanations

    special characters

    np_max-act · gemini-2.0-flash

    The main thing this neuron does is detect rare or nonstandard tokens—especially unusual symbols, foreign-language fragments, or garbled/malformed character sequences.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     textbook
    -0.06
     certification
    -0.06
    ::::/
    -0.06
     nồi
    -0.06
    (tweet
    -0.06
     丶
    -0.06
     atomic
    -0.06
     theology
    -0.06
     Interaction
    -0.06
     allergy
    -0.06
    POSITIVE LOGITS
    ��
    0.08
    mur
    0.08
     �
    0.08
    ha
    0.07
    �t
    0.07
    ে
    0.07
     �
    0.07
    .Exp
    0.06
     �
    0.06
    edia
    0.06
    Act Density 0.144%

    No Known Activations