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

    Neuronpedia

    INDEX
    Explanations

    references to the term "high" and its various contexts or derivatives

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

    high quality, speed, value descriptors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     betweenstory
    -0.54
     graça
    -0.48
    StructEnd
    -0.42
    retudo
    -0.42
     篇
    -0.42
     novità
    -0.41
     diferenças
    -0.41
    一方
    -0.40
     prática
    -0.39
     duração
    -0.39
    POSITIVE LOGITS
    high
    0.74
     high
    0.72
    High
    0.70
    HIGH
    0.70
     High
    0.68
    Highest
    0.65
     HIGH
    0.64
    highest
    0.63
     low
    0.60
     Highest
    0.59
    Act Density 0.066%

    No Known Activations