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

    Neuronpedia

    INDEX
    Explanations

    preparation

    np_max-act · gemini-2.0-flash

    narratives involving romantic encounters and relationships.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron fires on tokens describing preparatory or setup actions (e.g. “preparing,” “making,” “ensures,” “sets”) in a narrative.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    daughter
    -0.07
    Body
    -0.07
    _suffix
    -0.06
     khóa
    -0.06
    indexed
    -0.06
     dropping
    -0.06
    -un
    -0.06
    ,id
    -0.06
     evaluations
    -0.05
     �
    -0.05
    POSITIVE LOGITS
     كيل
    0.07
    *g
    0.07
    .CompilerServices
    0.07
    501
    0.06
    ètres
    0.06
    DBNull
    0.06
    िछ
    0.06
     disastr
    0.06
    ريس
    0.06
    ffb
    0.06
    Act Density 0.037%

    No Known Activations