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

    Neuronpedia

    INDEX
    Explanations

    phrases or words related to general concepts or topics

    oai_token-act-pair · gpt-3.5-turbo

    references to "things" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ramids
    -0.83
    aram
    -0.78
    omas
    -0.71
     quote
    -0.68
    idav
    -0.68
    imens
    -0.67
    sole
    -0.67
    edia
    -0.67
    pling
    -0.67
     indictment
    -0.67
    POSITIVE LOGITS
     imaginable
    1.00
     pertaining
    0.92
     transpired
    0.91
     relating
    0.74
     happening
    0.70
     whatsoever
    0.68
     happened
    0.68
     mathemat
    0.67
    THING
    0.67
     revolves
    0.67
    Act Density 0.040%

    No Known Activations