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

    Neuronpedia

    INDEX
    Explanations

    the number "3" as an important token in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    MigrationBuilder
    -0.90
    DoubleQuotes
    -0.78
    — 
    -0.77
     كومونز
    -0.70
    migrationBuilder
    -0.66
    saraba
    -0.62
    ImageContext
    -0.60
    addGap
    -0.60
    InjectAttribute
    -0.57
     charbon
    -0.57
    POSITIVE LOGITS
    1
    0.62
    7
    0.58
    5
    0.57
    4
    0.56
    2
    0.55
    9
    0.55
    3
    0.55
    6
    0.53
    8
    0.52
    0
    0.50
    Act Density 0.265%

    No Known Activations