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

    Neuronpedia

    INDEX
    Explanations

    conjunctions

    np_max-act · gemini-2.0-flash

    This neuron never produces a nonzero response—it doesn’t actually detect any pattern (i.e. it’s effectively “dead”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     talking
    -0.07
     laws
    -0.07
     lake
    -0.07
     heels
    -0.07
    _Normal
    -0.06
     маль
    -0.06
     topics
    -0.06
     Jones
    -0.06
     filter
    -0.06
     таблиц
    -0.06
    POSITIVE LOGITS
    (os
    0.07
    �
    0.07
     hesitate
    0.07
    \',
    0.06
    amilia
    0.06
    ình
    0.06
     يج
    0.06
     bonus
    0.06
    swick
    0.06
    ारक
    0.06
    Act Density 0.080%

    No Known Activations