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

    Neuronpedia

    INDEX
    Explanations

    potentially controversial actions or events related to public sentiment and debate

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

    verbs in past tense that indicate completed actions, particularly when followed by prepositions.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    words and phrases related to emotional pain or suffering.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi
    New Auto-Interp
    Negative Logits
    attuale
    -0.80
     pleaſure
    -0.77
     tomorrow
    -0.76
     derzeit
    -0.72
     theſe
    -0.72
     attualmente
    -0.72
     Monfieur
    -0.71
     künftig
    -0.68
     now
    -0.68
     actuelles
    -0.67
    POSITIVE LOGITS
     didn
    0.94
     damals
    0.92
    Wasn
    0.91
     did
    0.91
     was
    0.91
     went
    0.90
     wasn
    0.89
    didn
    0.88
     unknownFields
    0.85
     Wasn
    0.83
    Act Density 5.729%

    No Known Activations