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

    Neuronpedia

    INDEX
    Explanations

    transmit

    np_max-act · gemini-2.0-flash

    This neuron activates on terms describing radiation or X-ray penetration—i.e. tokens indicating beams passing through or being transmitted by a subject.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     nulla
    -0.07
     Başkan
    -0.07
    閉
    -0.07
    Cleaning
    -0.06
    ject
    -0.06
    Це
    -0.06
    �
    -0.06
     �
    -0.06
     Πα
    -0.06
    Jud
    -0.06
    POSITIVE LOGITS
    )<=
    0.07
     TIMES
    0.06
     与
    0.06
    μον
    0.06
    /(?
    0.06
    iscal
    0.06
     dropout
    0.06
     проп
    0.06
    写
    0.06
    assertFalse
    0.06
    Act Density 0.008%

    No Known Activations