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

    Neuronpedia

    INDEX
    Explanations

    code, errors, refunds

    np_acts-logits-general · gemini-2.5-flash-lite

    structural boundary/termination cues indicating the end of a statement, block, or conversation turn.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
     tattoos
    0.46
     esters
    0.42
     bonnes
    0.42
     proverbial
    0.40
    geqq
    0.40
    wpilibj
    0.40
    Callories
    0.39
     napp
    0.39
     sillons
    0.39
     vitamins
    0.38
    POSITIVE LOGITS
     confusing
    0.50
     संपादित
    0.47
    code
    0.44
     दरअसल
    0.42
     என்னுடைய
    0.42
    Xiaomi
    0.41
    Miami
    0.41
     Regex
    0.41
    https
    0.40
    In
    0.40
    Act Density 0.419%

    No Known Activations