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

    Neuronpedia

    INDEX
    Explanations

    practice

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric tokens with decimal points (floating‐point numbers).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     endowed
    -0.07
     Nob
    -0.07
     bordered
    -0.06
     stringBuffer
    -0.06
     goog
    -0.06
     emotional
    -0.06
     symb
    -0.06
     Elliot
    -0.06
     oxygen
    -0.06
    agonal
    -0.06
    POSITIVE LOGITS
     practice
    0.13
     practices
    0.11
     Practice
    0.11
    Practice
    0.08
     practise
    0.08
     Practices
    0.08
     Match
    0.08
    习
    0.08
     Larry
    0.08
     Pract
    0.08
    Act Density 0.029%

    No Known Activations