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

    Neuronpedia

    INDEX
    Explanations

    attends to various forms of "spread" from tokens related to applications or consequences mentioned later

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.72
    1:0.01
    2:0.01
    3:0.01
    4:0.15
    5:0.02
    6:0.01
    7:0.03
    Negative Logits
     houſe
    -0.47
     purpoſe
    -0.47
     itſelf
    -0.43
     raiſ
    -0.42
     Majefty
    -0.41
     ſtate
    -0.41
     Diſ
    -0.41
    ſelves
    -0.40
    TestingModule
    -0.40
     myſelf
    -0.39
    POSITIVE LOGITS
    </tfoot>
    0.40
    ANTLR
    0.34
    Tembelea
    0.33
    `;
    0.31
     formulate
    0.30
     veritable
    0.29
     linkovi
    0.29
    */,
    0.29
     illuminate
    0.28
    ';
    0.28
    Act Density 0.007%

    No Known Activations