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

    Neuronpedia

    INDEX
    Explanations

    formulas and facts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ായിക
    0.44
    वेल
    0.43
     സാധാരണ
    0.42
     substr
    0.40
     virtual
    0.40
    temperatur
    0.40
     thinking
    0.40
    』（
    0.40
     runaway
    0.39
     සා
    0.38
    POSITIVE LOGITS
    engono
    0.49
    Dark
    0.43
    awsze
    0.43
    opold
    0.39
     almeno
    0.38
    க்கால்
    0.38
    ейчас
    0.38
     peces
    0.37
    ตาล
    0.37
    멋
    0.37
    Act Density 0.002%

    No Known Activations