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

    Neuronpedia

    INDEX
    Explanations

    phrases related to persuasion or strong personal beliefs

    oai_token-act-pair · gpt-3.5-turbo

    phrases emphasizing trust and belief in statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    entin
    -0.87
     sidel
    -0.73
    imer
    -0.72
    ynski
    -0.72
    Thumbnail
    -0.72
    ija
    -0.70
    uj
    -0.68
    idian
    -0.66
    atlantic
    -0.64
    eners
    -0.64
    POSITIVE LOGITS
     sincerity
    0.73
     passionately
    0.68
     uncond
    0.65
     passion
    0.65
     hype
    0.64
     Dare
    0.62
    bly
    0.61
     courageous
    0.59
    acci
    0.59
     implicitly
    0.59
    Act Density 0.060%

    No Known Activations