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

    Neuronpedia

    INDEX
    Explanations

    sequences of punctuation or formatting marks at the end of sentences

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

    Punctuation after sentences or clauses

    np_acts-logits-general · gemini-2.0-flash

    sentence endings and starts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <em>
    -0.80
    <u>
    -0.74
    <code>
    -0.66
    <strong>
    -0.60
    ",$
    -0.59
    <i>
    -0.59
    {
    -0.58
    <s>
    -0.56
    -0.55
    )，
    -0.52
    POSITIVE LOGITS
     •
    1.10
     ==
    0.97
     *
    0.92
     >
    0.85
     -----------
    0.85
     ©
    0.85
     ----------
    0.84
     ==========
    0.81
     -----
    0.80
     --
    0.80
    Act Density 0.354%

    No Known Activations