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

    Neuronpedia

    INDEX
    Explanations

    emotional expressions and variations of feelings

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

    following "a", "the", "somewhat", or "energetically"

    np_acts-logits-general · gemini-2.0-flash

    surprise, disappointment, regret

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     للاسماء
    -0.42
     fédé
    -0.38
    anything
    -0.35
     féd
    -0.34
    ことができました
    -0.33
     Bagian
    -0.32
    izoen
    -0.31
    keyColumn
    -0.31
    EndInit
    -0.30
    <?,
    -0.30
    POSITIVE LOGITS
     surprise
    0.64
     surprised
    0.61
     shame
    0.61
     disappointment
    0.60
     ironic
    0.59
     disappointing
    0.58
     regret
    0.56
     disappointed
    0.56
     sad
    0.55
     dilemma
    0.55
    Act Density 0.386%

    No Known Activations