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

    Neuronpedia

    INDEX
    Explanations

    data related to demographic statistics and educational qualifications

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

    periods and punctuation marks that end sentences or clauses in statistical and research documents.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Negative Logits
    <unused17>
    -0.90
    <unused43>
    -0.90
    <unused42>
    -0.90
    <unused14>
    -0.89
    <unused74>
    -0.89
    <unused3>
    -0.89
    <unused23>
    -0.89
    <unused41>
    -0.89
    [@BOS@]
    -0.89
    <unused8>
    -0.89
    POSITIVE LOGITS
     nearly
    1.23
     Nearly
    1.13
     approximately
    1.12
    Nearly
    1.10
     Approximately
    1.05
    Approximately
    1.04
     half
    1.01
     almost
    0.99
    nearly
    0.93
     roughly
    0.92
    Act Density 1.160%

    No Known Activations