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

    Neuronpedia

    INDEX
    Explanations

    end of formulas

    np_max-act · gemini-2.0-flash

    This neuron detects high-precision decimal numbers (floating-point numerical tokens).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    brown
    -0.06
     irradi
    -0.06
     pall
    -0.06
     ])->
    -0.06
     nurses
    -0.06
    idy
    -0.06
    ye
    -0.06
    }()↵↵
    -0.06
    ір
    -0.06
    arro
    -0.06
    POSITIVE LOGITS
     sayıda
    0.07
    atoes
    0.07
     Forms
    0.07
     automat
    0.07
    housing
    0.07
     deported
    0.06
     دیگری
    0.06
    iplinary
    0.06
     يش
    0.06
    CAA
    0.06
    Act Density 0.003%

    No Known Activations