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

    Neuronpedia

    INDEX
    Explanations

    adjectives that describe qualities, particularly those that imply extremities or indescribable characteristics

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

    words ending in -able or -ible

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    findpost
    -0.83
    -0.81
    parsedMessage
    -0.81
    sizeCache
    -0.77
     Normdatei
    -0.72
    AddTagHelper
    -0.67
    Personendaten
    -0.65
    IVEREF
    -0.65
    DockStyle
    -0.63
    Tembelea
    -0.62
    POSITIVE LOGITS
     unimaginable
    0.48
     inconceivable
    0.43
     indescribable
    0.41
     Preußen
    0.40
     impossibile
    0.39
     great
    0.37
    Impossible
    0.37
     would
    0.36
     Impossible
    0.35
     unthinkable
    0.35
    Act Density 0.027%

    No Known Activations