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

    Neuronpedia

    INDEX
    Explanations

    format specifiers used in programming languages, particularly in input/output functions

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

    percentage or integer parsing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     MainAxisSize
    -0.83
    parsedMessage
    -0.79
     queſta
    -0.76
     $_"
    -0.75
    providedIn
    -0.75
    <unused8>
    -0.75
    <unused41>
    -0.75
    <unused52>
    -0.75
    <unused28>
    -0.75
    <unused14>
    -0.75
    POSITIVE LOGITS
     Schwartz
    0.33
    <eos>
    0.32
     forward
    0.31
     reason
    0.30
    ↵
    0.30
     heart
    0.29
     hesitation
    0.29
     Luce
    0.29
     Heart
    0.29
     entran
    0.29
    Act Density 0.001%

    No Known Activations