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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating degrees or types of classification

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

    "Kind of" or "sort of"

    np_acts-logits-general · gemini-2.0-flash

    kind of or sort of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     absolutamente
    -0.61
     africaine
    -0.56
     absolutely
    -0.56
     assolutamente
    -0.54
     asla
    -0.52
     absolument
    -0.51
     exceedingly
    -0.49
     apparently
    -0.49
     milioane
    -0.49
     zoveel
    -0.48
    POSITIVE LOGITS
     sorta
    0.92
     оригіналу
    0.88
     unknownFields
    0.85
    Kinda
    0.77
    +#+
    0.75
     متعلقه
    0.75
     Kinda
    0.74
    EndInit
    0.74
    matchCondition
    0.74
     referenties
    0.73
    Act Density 0.150%

    No Known Activations