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

    Neuronpedia

    INDEX
    Explanations

    acronyms

    np_max-act · gemini-2.0-flash

    The neuron responds strongly to single uppercase letters (and their surrounding punctuation) when they’re being used as parts of acronyms.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     hug
    -0.07
    afil
    -0.07
     bab
    -0.07
     hugs
    -0.06
     fing
    -0.06
    Ars
    -0.06
    .sig
    -0.06
    Зап
    -0.06
     anymore
    -0.06
     advises
    -0.06
    POSITIVE LOGITS
     ked
    0.07
    izzazione
    0.07
    uelve
    0.07
    PE
    0.07
    verts
    0.06
     بخشی
    0.06
     blessing
    0.06
     programm
    0.06
    (hr
    0.06
    ifier
    0.06
    Act Density 0.013%

    No Known Activations