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

    Neuronpedia

    INDEX
    Explanations

    patterns related to structured data or parameters, particularly those with underscore-prefixed terms

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

    shows

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    原始内容存档于
    -0.38
     onCancelled
    -0.29
    <bos>
    -0.29
     又
    -0.27
     zudem
    -0.26
     hingga
    -0.26
    しかも
    -0.26
     disfr
    -0.25
     насељу
    -0.24
     namun
    -0.24
    POSITIVE LOGITS
    <unused28>
    0.75
    <unused79>
    0.75
    <unused23>
    0.74
    <unused17>
    0.74
    <unused68>
    0.74
    <unused74>
    0.74
    <unused16>
    0.74
    <unused8>
    0.74
    <unused3>
    0.74
    [@BOS@]
    0.74
    Act Density 0.191%

    No Known Activations