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

    Neuronpedia

    INDEX
    Explanations

    }"

    np_max-act · gemini-2.0-flash

    The neuron flags tokens that are part of figure or illustration references (e.g. “Figure,” “Fig.,” section numbers, bracketed or parenthesized figure labels).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ied
    -0.07
    anda
    -0.06
    esture
    -0.06
    ICH
    -0.06
    GLOSS
    -0.06
     Mixed
    -0.06
    Moves
    -0.06
     Sizes
    -0.06
    imated
    -0.06
     bec
    -0.06
    POSITIVE LOGITS
     inflater
    0.07
    ","+
    0.07
    uito
    0.07
     protože
    0.06
     Rel
    0.06
     بالأ
    0.06
     pz
    0.06
     ghetto
    0.06
    PostalCodesNL
    0.06
    ={{↵
    0.06
    Act Density 0.008%

    No Known Activations