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

    Neuronpedia

    INDEX
    Explanations

    expressions of apology and requests for forgiveness

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

    apologize, apology, apologies

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     MutableLiveData
    -0.46
    IntoConstraints
    -0.43
    PreferredItem
    -0.42
     SafeMath
    -0.40
     Picchu
    -0.39
    ImageContext
    -0.39
    ftant
    -0.39
    Czytaj
    -0.39
     keduanya
    -0.39
     BoxFit
    -0.38
    POSITIVE LOGITS
     apologized
    0.93
     apologizing
    0.93
     apology
    0.89
     apologised
    0.86
     SORRY
    0.85
     sorry
    0.84
     apologize
    0.81
     apologies
    0.80
     apologe
    0.78
     apologise
    0.75
    Act Density 0.012%

    No Known Activations