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

    Neuronpedia

    INDEX
    Explanations

    This neuron appears inactive in these examples — it does not respond to any tokens (a dead or unused neuron).

    oai_token-act-pair · gpt-5-miniTriggered by @chenshw0109
    New Auto-Interp
    Negative Logits
    <eos>
    -0.87
    AnchorStyles
    -0.56
    jgl
    -0.55
    |()
    -0.54
    aft
    -0.49
    jooq
    -0.48
     poste
    -0.46
     فريبيس
    -0.46
     ||
    -0.45
    AutoScaleMode
    -0.45
    POSITIVE LOGITS
     Efq
    0.89
     againſt
    0.88
     whoſe
    0.86
     bonté
    0.82
     myſelf
    0.82
     himſelf
    0.80
    ſelf
    0.80
     itſelf
    0.80
     tillbaka
    0.79
     Jefus
    0.79
    Act Density 0.000%

    No Known Activations

    This feature has no known activations.