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

    Neuronpedia

    INDEX
    Explanations

    attends to numeric values from preceding tokens

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.12
    1:0.14
    2:0.12
    3:0.14
    4:0.11
    5:0.09
    6:0.11
    7:0.13
    Negative Logits
    ArrowToggle
    -0.40
    ();)
    -0.40
    ;”
    -0.37
     étoient
    -0.36
    Abp
    -0.36
     africains
    -0.36
     avoient
    -0.35
    ์ตูน
    -0.35
     فريبيس
    -0.35
     quæ
    -0.34
    POSITIVE LOGITS
     ResponseEntity
    0.36
    ENAME
    0.35
    eeeeee
    0.33
    enken
    0.32
     Mo
    0.32
     defStyleAttr
    0.31
    rtype
    0.30
     Po
    0.30
     Moos
    0.30
    genommen
    0.29
    Act Density 0.032%

    No Known Activations