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

    Neuronpedia

    INDEX
    Explanations

    numerical values and their associated units in scientific contexts

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

    Percentages or weights within a composition

    np_acts-logits-general · gemini-2.0-flash

    percentages and compositions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    mergeFrom
    -0.65
     дописавши
    -0.64
    ագրություններ
    -0.61
    Personensuche
    -0.56
     reutiliz
    -0.54
    alquiler
    -0.54
    aliyun
    -0.52
     Baillargeon
    -0.51
    +][
    -0.51
     Pingback
    -0.50
    POSITIVE LOGITS
     content
    0.82
     percentage
    0.80
     Composition
    0.80
     composition
    0.78
     percent
    0.78
    Percent
    0.75
    composition
    0.75
     (%
    0.74
     percentages
    0.73
    content
    0.72
    Act Density 0.463%

    No Known Activations