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

    Neuronpedia

    INDEX
    Explanations

    words indicating significant quantities or extremes

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

    large amounts and ratios

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    DCA
    -0.54
    ("")]
    -0.46
    SCE
    -0.43
    __":
    -0.43
     MTL
    -0.42
     DCA
    -0.41
     createContext
    -0.39
     æ
    -0.39
     Ill
    -0.39
    CompilerServices
    -0.38
    POSITIVE LOGITS
     whopping
    2.03
     huge
    0.77
     énorme
    0.75
     staggering
    0.74
     yüzde
    0.70
    huge
    0.68
     enormes
    0.68
    なんと
    0.68
     HUGE
    0.67
    Huge
    0.66
    Act Density 0.001%

    No Known Activations