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

    Neuronpedia

    INDEX
    Explanations

    words related to extraction

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "ext" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Shogun
    -0.67
     Garland
    -0.67
     Spur
    -0.63
     Fields
    -0.62
     Knights
    -0.62
     Siege
    -0.60
     footing
    -0.60
     Swift
    -0.59
    folk
    -0.59
     poisoning
    -0.57
    POSITIVE LOGITS
    ention
    1.19
    ensible
    1.08
    ricular
    1.06
    orting
    1.02
    entric
    1.02
    ensive
    1.01
    oll
    0.98
    orted
    0.98
    ingu
    0.98
    ensional
    0.97
    Act Density 0.021%

    No Known Activations