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

    Neuronpedia

    INDEX
    Explanations

    numeric figures and measurements

    oai_token-act-pair · gpt-3.5-turbo

    conjunctions and the presence of the word "and" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    èª
    -0.72
    SpaceEngineers
    -0.72
    cture
    -0.64
    endif
    -0.60
    cially
    -0.59
    dayName
    -0.59
    hee
    -0.59
    rather
    -0.58
    Anyway
    -0.58
    EMOTE
    -0.57
    POSITIVE LOGITS
     150
    1.23
     250
    1.18
     144
    1.18
     120
    1.17
     140
    1.17
     132
    1.17
     160
    1.16
     170
    1.15
     176
    1.14
     175
    1.14
    Act Density 0.219%

    No Known Activations