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

    Neuronpedia

    INDEX
    Explanations

    variations and measurements related to levels in different contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    level of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     NSCoder
    -0.45
    XmlAttribute
    -0.42
     compressed
    -0.42
     Boston
    -0.42
     cracks
    -0.42
     shoe
    -0.41
     fastest
    -0.40
     PyObject
    -0.40
     Columbus
    -0.40
     Dör
    -0.40
    POSITIVE LOGITS
     levels
    1.07
     poziom
    1.04
    levels
    1.00
    level
    0.98
     level
    0.96
     nivelul
    0.96
     niveles
    0.94
     Levels
    0.94
     LEVELS
    0.93
     níveis
    0.93
    Act Density 0.079%

    No Known Activations