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

    Neuronpedia

    INDEX
    Explanations

    superlatives or words related to extremity

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "most" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    heid
    -0.74
    rompt
    -0.74
    pload
    -0.72
    æ©
    -0.71
    oak
    -0.67
     Films
    -0.64
     Recording
    -0.63
    ategories
    -0.63
     Oaks
    -0.62
    instead
    -0.61
    POSITIVE LOGITS
     important
    0.98
    likely
    0.91
     obvious
    0.90
     plausible
    0.89
     likely
    0.87
     efficient
    0.87
     profitable
    0.85
     importantly
    0.84
     noticeable
    0.84
     prominent
    0.84
    Act Density 0.058%

    No Known Activations