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

    Neuronpedia

    INDEX
    Explanations

    tokens that are numeric timestamps or high-precision numeric date/time markers.

    oai_token-act-pair · gpt-5-miniTriggered by @yooniel31
    New Auto-Interp
    Negative Logits
    不需要
    0.43
     indire
    0.43
     back
    0.41
     něk
    0.39
     rout
    0.36
     unaware
    0.36
     mixes
    0.36
    不能
    0.36
     doesn
    0.35
    ক্ষের
    0.35
    POSITIVE LOGITS
     onwards
    0.57
     tarihinde
    0.56
     silam
    0.56
     թվական
    0.54
    년
    0.54
     pomeriggio
    0.54
    zeptember
    0.53
     সালের
    0.50
    時点
    0.49
     Πολ
    0.49
    Act Density 0.016%

    No Known Activations