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

    Neuronpedia

    INDEX
    Explanations

    "bad" and "ugly"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     bacilli
    0.50
     breakfasts
    0.49
     immunoprec
    0.48
     assemblies
    0.47
     desserts
    0.47
     painkillers
    0.47
     egos
    0.46
     analges
    0.46
     syndromes
    0.46
    ították
    0.46
    POSITIVE LOGITS
    4
    0.52
    1
    0.49
    3
    0.48
     C
    0.47
    0.46
    ただし
    0.46
    2
    0.46
     +
    0.45
     Answer
    0.45
    5
    0.44
    Act Density 0.002%

    No Known Activations