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

    Neuronpedia

    INDEX
    Explanations

    letters

    np_max-act · gemini-2.0-flash

    This neuron detects phrases that specify “N letters” (e.g. “two letters,” “three letters,” “four letters”) in the context of a probability or replacement description.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Acceleration
    -0.06
    np
    -0.06
     fancy
    -0.06
    галі
    -0.06
    enn
    -0.06
     paed
    -0.06
    胡
    -0.06
     Shawn
    -0.06
     правда
    -0.06
     ↔
    -0.05
    POSITIVE LOGITS
     Module
    0.07
    .timeScale
    0.07
     الكتاب
    0.07
    segment
    0.06
    Package
    0.06
     firearm
    0.06
     ผล
    0.06
    >'
↵
    0.06
    	initialize
    0.06
    liable
    0.06
    Act Density 0.001%

    No Known Activations