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

    Neuronpedia

    INDEX
    Explanations

    references to "round" and related terms indicating numerical adjustments

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

    rounding instructions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     HasFactory
    -0.61
     []:
    -0.51
    CodeAttribute
    -0.46
     <<<<<<<<<<<<<<
    -0.46
    ogaster
    -0.45
    DatabaseError
    -0.43
    erweise
    -0.42
     mukana
    -0.42
     gezegd
    -0.42
    MenuInflater
    -0.42
    POSITIVE LOGITS
    Round
    2.31
     Round
    2.28
    round
    2.20
    ROUND
    1.97
     ROUND
    1.91
     round
    1.65
    rounds
    1.65
     Rounds
    1.62
    Rounds
    1.52
     rounds
    1.45
    Act Density 0.005%

    No Known Activations