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

    Neuronpedia

    INDEX
    Explanations

    Code and chat snippets

    np_max-act · gemini-2.0-flash

    This neuron selectively activates on floating‐point numeric tokens (decimal numbers).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .translate
    -0.07
    AV
    -0.07
    throws
    -0.07
    itable
    -0.06
    telephone
    -0.06
    ря
    -0.06
     کنم
    -0.06
    .Prop
    -0.06
    ặ
    -0.06
    ()(
    -0.06
    POSITIVE LOGITS
    间
    0.07
    ความร
    0.06
    �
    0.06
     Augustine
    0.06
    679
    0.06
     Şah
    0.06
    Ann
    0.06
    \models
    0.06
    νη
    0.06
    ━━━━━━━━
    0.06
    Act Density 0.026%

    No Known Activations