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

    Neuronpedia

    INDEX
    Explanations

    next

    np_max-act · gemini-2.0-flash

    The neuron is sensitive to step‐by‐step transition cues—words like “next,” “previous,” “continuing,” and similar that signal the progression of sequential reasoning.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ême
    -0.08
     Hoffman
    -0.07
    occan
    -0.07
     Expand
    -0.07
    _nb
    -0.07
     Dia
    -0.07
     harus
    -0.06
    紙
    -0.06
     lamps
    -0.06
    .si
    -0.06
    POSITIVE LOGITS
    (CONFIG
    0.07
     Shaw
    0.07
    _release
    0.06
     результат
    0.06
     compromises
    0.06
     tract
    0.06
     Rocky
    0.06
     Lindsey
    0.06
     footing
    0.06
     Shiv
    0.06
    Act Density 0.012%

    No Known Activations