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

    Neuronpedia

    INDEX
    Explanations

    Reinvested dividends

    np_max-act · gemini-2.0-flash

    This neuron detects the subword pattern corresponding to the word “reinvested” (i.e. the “rein – vest – ed” token sequence).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    aru
    -0.07
     IID
    -0.07
    irie
    -0.06
    айд
    -0.06
     DE
    -0.06
    _tem
    -0.06
    خراج
    -0.06
    .Fail
    -0.06
    (Equal
    -0.06
    011
    -0.06
    POSITIVE LOGITS
     blindly
    0.07
    (stdout
    0.06
     δημιουργ
    0.06
     diets
    0.06
     gifts
    0.06
     binds
    0.06
     Thời
    0.06
     �
    0.06
     allocating
    0.06
    ̂
    0.06
    Act Density 0.002%

    No Known Activations