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

    Neuronpedia

    INDEX
    Explanations

    mentions of layers or levels with high emphasis or importance

    oai_token-act-pair · gpt-3.5-turbo

    references to layers in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ãĥīãĥ©
    -0.84
     Predators
    -0.75
    speaking
    -0.74
     STAR
    -0.72
    orie
    -0.71
    date
    -0.70
    lishing
    -0.70
     FRE
    -0.66
    itarian
    -0.66
     Rao
    -0.65
    POSITIVE LOGITS
     layers
    1.35
     layer
    1.21
     thickness
    0.99
    layer
    0.91
    Layer
    0.87
     Layer
    0.83
     opacity
    0.80
    coat
    0.79
     coats
    0.77
     veil
    0.77
    Act Density 0.019%

    No Known Activations