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

    Neuronpedia

    INDEX
    Explanations

    technical terms or concepts

    oai_token-act-pair · gpt-3.5-turbo

    references to technical aspects or qualities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Sheen
    -0.77
    xon
    -0.75
    inas
    -0.74
     Cull
    -0.72
     istg
    -0.72
    vic
    -0.72
     largeDownload
    -0.71
     Graves
    -0.70
     Cele
    -0.66
    Shar
    -0.66
    POSITIVE LOGITS
     technical
    1.05
    technical
    1.02
     jargon
    0.93
     glitches
    0.91
     specifications
    0.90
     difficulties
    0.83
     proficiency
    0.83
    idad
    0.82
     sophistication
    0.81
    smanship
    0.81
    Act Density 0.007%

    No Known Activations