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

    Neuronpedia

    INDEX
    Explanations

    specific phrases that validate or reinforce previous information or beliefs

    oai_token-act-pair · gpt-3.5-turbo

    phrases that confirm or support previous information or findings

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Became
    -0.73
     Ended
    -0.73
    abouts
    -0.72
    abilia
    -0.64
     interacts
    -0.63
    illion
    -0.63
     Swordsman
    -0.62
     compensated
    -0.61
    usercontent
    -0.61
     resembled
    -0.61
    POSITIVE LOGITS
     sentiments
    1.10
     trend
    1.07
     notion
    1.00
     longstanding
    0.97
     tradition
    0.97
     sentiment
    0.94
     suspicions
    0.94
     assumption
    0.93
     mantra
    0.91
     recommendations
    0.91
    Act Density 0.435%

    No Known Activations