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

    Neuronpedia

    INDEX
    Explanations

    the word "completely."

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

    completely + descriptor

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Inscrivez
    -0.58
     Speise
    -0.52
     الرياضيه
    -0.52
     trône
    -0.51
     äta
    -0.48
    SerializeField
    -0.48
    Datuak
    -0.47
     goût
    -0.46
     africains
    -0.46
     enfans
    -0.45
    POSITIVE LOGITS
     completely
    1.20
     Completely
    1.09
     totally
    1.05
    Completely
    1.05
    completely
    1.03
     entirely
    0.95
    totally
    0.93
     utterly
    0.92
     Totally
    0.91
    Totally
    0.86
    Act Density 0.010%

    No Known Activations