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

    Neuronpedia

    INDEX
    Explanations

    symbols and markers typically associated with data formatting or code syntax

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

    Tokens followed by digits

    np_acts-logits-general · gemini-2.0-flash

    numerical sequences like `code` 4

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Four
    -1.67
    Fourth
    -1.65
     Fourth
    -1.64
    Four
    -1.62
    FOUR
    -1.60
     four
    -1.56
    four
    -1.55
    Fifth
    -1.54
     Five
    -1.51
     Fifth
    -1.49
    POSITIVE LOGITS
     industriels
    0.48
     virke
    0.46
    atkan
    0.43
    riez
    0.41
     EIR
    0.40
     succede
    0.39
     nguyên
    0.39
    níci
    0.38
    leitung
    0.38
     تب
    0.38
    Act Density 0.338%

    No Known Activations