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

    Neuronpedia

    INDEX
    Explanations

    adjectives or adverbs denoting the highest degree or superlative form of an attribute

    oai_token-act-pair · gpt-3.5-turbo

    phrases that emphasize the word "most" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Films
    -0.71
    heid
    -0.67
     Nero
    -0.60
     Fist
    -0.60
    rompt
    -0.59
    instead
    -0.59
     Tam
    -0.57
    onut
    -0.56
     Reloaded
    -0.56
     Creed
    -0.56
    POSITIVE LOGITS
     importantly
    0.99
    likely
    0.98
     likely
    0.88
     prominently
    0.82
    mos
    0.81
     notably
    0.78
     important
    0.77
    afa
    0.75
     obvious
    0.74
    OWER
    0.73
    Act Density 0.064%

    No Known Activations