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

    Neuronpedia

    INDEX
    Explanations

    requests for forgiveness and expressions of remorse

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

    Expressions of apology or requests for forgiveness

    np_acts-logits-general · gemini-2.0-flash

    asking for forgiveness

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    endencia
    -0.47
    onicus
    -0.46
    gypti
    -0.46
     suspiciously
    -0.46
     tanong
    -0.44
     inconnu
    -0.44
    ktır
    -0.44
     Wilcox
    -0.43
     XCTest
    -0.43
    cination
    -0.42
    POSITIVE LOGITS
     apology
    2.52
     apologize
    2.45
     apologized
    2.42
     apologizing
    2.38
     apologies
    2.36
     apologise
    2.20
     apolog
    2.19
     apologised
    2.17
     sorry
    1.90
     apologe
    1.82
    Act Density 0.334%

    No Known Activations