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

    Neuronpedia

    INDEX
    Explanations

    ytypical

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on the sequence “typical” (as in “atypical” or “typical”) in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    )("
    -0.07
     fishes
    -0.06
    ût
    -0.06
    nov
    -0.06
    suggest
    -0.06
     jego
    -0.06
    rebbe
    -0.06
     необхідно
    -0.06
    <const
    -0.06
    Evaluate
    -0.06
    POSITIVE LOGITS
    091
    0.07
     ek
    0.07
    КА
    0.07
    "<?
    0.06
     الجن
    0.06
    елі
    0.06
     کمی
    0.06
     victory
    0.06
    hen
    0.06
    یک
    0.06
    Act Density 0.002%

    No Known Activations