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

    Neuronpedia

    INDEX
    Explanations

    tokens that occur at the start of a sentence or turn (beginning-of-sentence/turn tokens).

    oai_token-act-pair · gpt-5-miniTriggered by @yooniel31
    New Auto-Interp
    Negative Logits
    чан
    0.45
     moderators
    0.42
    पांच
    0.40
     frm
    0.38
     Sine
    0.38
     Kik
    0.38
    ച്ചി
    0.38
     shale
    0.38
     SOA
    0.38
    чне
    0.37
    POSITIVE LOGITS
    imizin
    0.40
    ımızı
    0.40
    acup
    0.39
     تربيع
    0.38
    Bit
    0.37
    IM
    0.36
    accouchement
    0.35
    imiz
    0.35
    ienes
    0.35
    ètent
    0.35
    Act Density 0.000%

    No Known Activations