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

    Neuronpedia

    INDEX
    Explanations

    concessive phrases indicating uncertainty or doubt about ideas

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

    Following "but"

    np_acts-logits-general · gemini-2.0-flash

    contrasting but leading to a summary

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     BoxDecoration
    -0.76
     незавершена
    -0.72
    yntaxException
    -0.64
    WireFormatLite
    -0.62
     mergeFrom
    -0.61
    Diwedd
    -0.59
    InputBorder
    -0.58
     Signalez
    -0.58
    EndContext
    -0.58
     AssemblyCompany
    -0.57
    POSITIVE LOGITS
    complexContent
    0.41
     generally
    0.33
     sérieux
    0.32
     основной
    0.31
     pretty
    0.30
    ⅘
    0.30
    SUMMARY
    0.29
     głów
    0.29
    mainMenu
    0.29
    Main
    0.29
    Act Density 0.277%

    No Known Activations