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

    Neuronpedia

    INDEX
    Explanations

    attends to numerical tokens related to categories from numerical tokens in the 21st century

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.03
    1:0.09
    2:0.04
    3:0.04
    4:0.17
    5:0.53
    6:0.03
    7:0.03
    Negative Logits
    ThroughAttribute
    -0.42
     onAnimation
    -0.40
    ScopeManager
    -0.38
     cesse
    -0.37
    AddTagHelper
    -0.37
     Infórmanos
    -0.36
    脚注の使い方
    -0.36
     Мексичка
    -0.36
    RepeatedField
    -0.34
    TagHelper
    -0.34
    POSITIVE LOGITS
    moveToFirst
    0.30
     human
    0.27
    Inggris
    0.26
     Colonna
    0.26
    perfusion
    0.25
     Mob
    0.25
    MCA
    0.25
    izza
    0.25
    Networks
    0.25
     Cities
    0.25
    Act Density 3.784%

    No Known Activations