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

    Neuronpedia

    INDEX
    Explanations

    instances of subtlety or nuanced expressions

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

    subtle changes and nuances

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     chargeur
    -0.61
     feroit
    -0.57
    chargez
    -0.55
     auroit
    -0.54
     ainfi
    -0.52
     enfans
    -0.52
     étoit
    -0.52
     pouvoit
    -0.52
    actéristi
    -0.51
     vœ
    -0.51
    POSITIVE LOGITS
    subtle
    1.10
     subtle
    1.09
     Subtle
    0.98
     subtly
    0.95
     sutil
    0.83
     subtlety
    0.76
     subtleties
    0.75
     sottile
    0.71
    微妙
    0.66
     subtil
    0.66
    Act Density 0.004%

    No Known Activations