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

    Neuronpedia

    INDEX
    Explanations

    terms indicating timeframes, specifically early and late stages

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    early or late

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    তথ্যসূত্র
    -0.60
     esternos
    -0.56
     fVar
    -0.50
     valla
    -0.50
     Elli
    -0.49
    reactstrap
    -0.49
    TRIBUN
    -0.47
    quations
    -0.47
    というもの
    -0.47
    Ссылки
    -0.47
    POSITIVE LOGITS
     early
    0.94
    early
    0.92
     late
    0.90
     LATE
    0.88
    late
    0.85
     Early
    0.85
    Early
    0.84
     Late
    0.84
     EARLY
    0.83
    EARLY
    0.82
    Act Density 0.106%

    No Known Activations