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

    Neuronpedia

    INDEX
    Explanations

    high-frequency words and pronouns typically used in discussions

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

    had been analyzed

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <eos>
    -0.28
     
    -0.28
    ↵
    -0.27
     wynosi
    -0.26
    ambilan
    -0.25
     również
    -0.25
     Grenze
    -0.24
     bluza
    -0.24
    ↵↵
    -0.23
     is
    -0.23
    POSITIVE LOGITS
    <unused41>
    1.00
    <unused79>
    0.99
    [@BOS@]
    0.99
    <unused43>
    0.99
    <unused52>
    0.99
    <unused28>
    0.99
    <unused68>
    0.99
    <unused74>
    0.99
    <unused14>
    0.99
    <unused23>
    0.99
    Act Density 0.157%

    No Known Activations