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

    Neuronpedia

    INDEX
    Explanations

    names of politicians or individuals

    oai_token-act-pair · gpt-3.5-turbo

    proper nouns and names

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Vessel
    -0.82
     Ow
    -0.74
     Gauntlet
    -0.73
     Cycle
    -0.72
     Gear
    -0.72
     Wat
    -0.71
     808
    -0.71
     TABLE
    -0.70
     Tab
    -0.70
     Torch
    -0.68
    POSITIVE LOGITS
    an
    1.44
    AN
    1.31
    ans
    1.31
    ano
    1.27
    ania
    1.22
    anian
    1.17
    ani
    1.17
    acan
    1.14
    anos
    1.13
    anism
    1.12
    Act Density 0.202%

    No Known Activations