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

    Neuronpedia

    INDEX
    Explanations

    prepositions and hyphenated phrases

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    ulhu
    -0.95
     Tycoon
    -0.73
     modification
    -0.72
     chant
    -0.72
     AAC
    -0.70
     Boone
    -0.68
     Dickinson
    -0.67
     consolidation
    -0.65
     Boys
    -0.64
     Doodle
    -0.63
    POSITIVE LOGITS
    average
    1.25
    expected
    1.19
    anticipated
    1.16
    mentioned
    1.15
    the
    1.15
    normal
    1.13
    usual
    1.08
    repl
    1.06
    strength
    1.06
    advert
    1.05
    Act Density 0.024%

    No Known Activations