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

    Neuronpedia

    INDEX
    Explanations

    code examples

    np_max-act · gemini-2.0-flash

    This neuron fires on words immediately followed by runs of two or more consecutive spaces.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     readers
    -0.07
     Development
    -0.06
     Baker
    -0.06
     Retreat
    -0.06
     adapt
    -0.06
     Dillon
    -0.06
     Booth
    -0.06
     reloc
    -0.06
    solid
    -0.06
     cancellationToken
    -0.06
    POSITIVE LOGITS
    &&(
    0.07
     moyen
    0.07
    'http
    0.06
    .userAgent
    0.06
     Chúa
    0.06
    ह
    0.06
     куп
    0.06
    Ball
    0.06
    uide
    0.06
    (金
    0.06
    Act Density 0.031%

    No Known Activations