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

    Neuronpedia

    INDEX
    Explanations

    attends to tokens indicating financial metrics or references from tokens indicating numerical values or comparisons

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.08
    1:0.11
    2:0.08
    3:0.11
    4:0.12
    5:0.06
    6:0.22
    7:0.18
    Negative Logits
    enumi
    -0.23
    PUS
    -0.22
    ítése
    -0.22
     sitten
    -0.21
    akers
    -0.20
    توض
    -0.20
    бут
    -0.20
    comer
    -0.19
     hinein
    -0.19
     TRA
    -0.19
    POSITIVE LOGITS
     Roskov
    0.46
    Życiorys
    0.38
     MainAxisSize
    0.37
     kasarigan
    0.36
    0.35
    TestingModule
    0.35
     Administrativna
    0.34
    NewUrlParser
    0.33
    InjectAttribute
    0.33
    AddHtmlAttribute
    0.33
    Act Density 0.016%

    No Known Activations