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

    Neuronpedia

    INDEX
    Explanations

    sentences expressing problems or issues related to experiences or software errors

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

    but / Now followed by a statement

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    RotationOrder
    -0.56
    RectangleBorder
    -0.52
     hiszen
    -0.50
    writeField
    -0.49
    tiéndose
    -0.48
    ArgumentParser
    -0.47
     Kindheit
    -0.47
    Willkommen
    -0.47
    那就
    -0.47
    \{\\
    -0.47
    POSITIVE LOGITS
     whenever
    0.87
     Whenever
    0.75
     everytime
    0.73
    whenever
    0.70
    Whenever
    0.70
     when
    0.66
     occasionally
    0.64
     intermittently
    0.60
     Occasionally
    0.59
     suddenly
    0.58
    Act Density 0.112%

    No Known Activations