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

    Neuronpedia

    INDEX
    Explanations

    abbreviations or terms related to weight measurement unit, particularly "BW."

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

    bw or BW followed by punctuation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     domicili
    -0.44
    tangentMode
    -0.43
     robusta
    -0.43
     Stol
    -0.42
    >())
    -0.41
    käse
    -0.41
    UObject
    -0.40
    hassee
    -0.40
     enfans
    -0.40
    Seeder
    -0.39
    POSITIVE LOGITS
    bw
    2.28
     BW
    2.20
    BW
    2.19
     bw
    2.03
    Bw
    1.55
     bandwidth
    0.82
     Bandwidth
    0.79
    bandwidth
    0.77
    Bandwidth
    0.76
    pw
    0.75
    Act Density 0.003%

    No Known Activations