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

    Neuronpedia

    INDEX
    Explanations

    generic text fragments

    np_max-act · gemini-2.0-flash

    This neuron fires on the opening words of new sections or sentences (i.e. the first few tokens right after a break).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    tell
    -0.07
     Mature
    -0.06
    ัช
    -0.06
     gradient
    -0.06
     graphene
    -0.06
    Accept
    -0.06
    ไฟ
    -0.06
     UIAlertController
    -0.06
     final
    -0.06
     Clock
    -0.06
    POSITIVE LOGITS
    sembl
    0.07
     порядку
    0.07
     случа
    0.07
    ForMember
    0.07
     gp
    0.07
     wyn
    0.07
    гот
    0.06
     легко
    0.06
     wykon
    0.06
    tracer
    0.06
    Act Density 0.083%

    No Known Activations