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

    Neuronpedia

    INDEX
    Explanations

    Nothing — this neuron remains inactive and does not detect any specific tokens or patterns.

    oai_token-act-pair · gpt-5-miniTriggered by @chenshw0109
    New Auto-Interp
    Negative Logits
    <eos>
    -1.23
    ↵↵
    -1.11
    ↵
    -0.88
     (
    -0.86
    <strong>
    -0.85
      
    -0.84
    i
    -0.83
    <em>
    -0.82
     
    -0.82
    .
    -0.81
    POSITIVE LOGITS
    ValueStyle
    2.17
     itſelf
    2.16
     myſelf
    2.14
    ^(@)
    2.09
    ſelves
    2.08
     Roskov
    1.97
    Personendaten
    1.95
    ſelf
    1.94
     doubtnut
    1.94
     ſind
    1.92
    Act Density 0.000%

    No Known Activations

    This feature has no known activations.