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

    Neuronpedia

    INDEX
    Explanations

    proper nouns, including locations and organizations

    oai_token-act-pair · gpt-3.5-turbo

    references to government or legal structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ocate
    -0.73
    ibraries
    -0.71
    ritis
    -0.71
    escription
    -0.65
    ogeneity
    -0.64
    ¶ħ
    -0.63
    atron
    -0.62
    ropri
    -0.62
    acons
    -0.61
    orically
    -0.61
    POSITIVE LOGITS
     basis
    1.41
     occasions
    1.28
     behalf
    1.22
     occasion
    1.12
     doorstep
    1.09
     grounds
    1.08
     bandwagon
    1.04
    imeo
    0.98
     sidelines
    0.96
     shelf
    0.96
    Act Density 0.719%

    No Known Activations