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

    Neuronpedia

    INDEX
    Explanations

    Punctuation and connector words

    np_max-act · gemini-2.0-flash

    The neuron selectively lights up on numeric tokens—especially decimal numbers—indicating it’s looking for floating‐point or other numeric values.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    -UA
    -0.07
    _CLOSED
    -0.07
     channels
    -0.07
     Зав
    -0.06
    #:
    -0.06
    Doctor
    -0.06
    _binary
    -0.06
    _WE
    -0.06
     Що
    -0.06
    .book
    -0.06
    POSITIVE LOGITS
     excessively
    0.07
    ')]
    0.06
    	os
    0.06
    ิเวณ
    0.06
    砂
    0.06
     getattr
    0.06
    ethylene
    0.06
     laisse
    0.06
    upportInitialize
    0.06
    εργ
    0.06
    Act Density 0.039%

    No Known Activations